View Full Version : Squaring and leveling table

01-04-2005, 11:36 PM
Hello all, my Shopbot is set to ship on the 14th of this month. I'm building my table from the steel plans on the website and just got it all assembled today. I have not yet squared it and leveled it yet though. My question is about the sides that the rails bolt to. The website plans say that the Overall width of the table should be 64 1/2". Mine, when measured at one end measures 64 1/4" and the opposite end measures just over 64 3/16". In the middle though the metal bows out and it measures 64 1/2". My question is this,is this acceptable for the rail to go on or does it need to me closer. I'm thinking that this will work as long as the rail has support under it and can be made straight.
Now for across the top of the beam. I assume this has to be pretty darn close to flat. But it seems like mine is pretty close to flat towards the ends but in the middle is out. I guess it is twisted some way. does this have to be perfectly flat all the way down the length or what is an exceptable tolerence? Or should I just wait on the machine to get here and get it all straight with the rails. Thanks for any info.

01-05-2005, 12:01 AM
You have done enough so far. Wait for the machine and then you will have a few more questions.

01-05-2005, 04:00 PM
Gerald, you are teasing the poor guy! Remember how excited we all were to get our machines set up? He's going to go nuts waiting if he isn't messing with something.

Wayne, I have to admit that Gerald is right. You can't really do much more until your machine gets there. You should know that when you bolt the X-rails to the C-channel that you will shim out minor deviations in Z and have enough clearance in the bolt holes to ignor some deviations in x & Y. The only thing I would recommend that you worry about while you wait is getting the top of both C-channel members as coplanar as possible (in Z). I actually had to spend some time on this when I built my table. There seemed to be some draft on the channel I used so I actually had to splay my legs outward a bit to get the top surfaces aligned.

01-30-2005, 02:10 AM
I just read your post, I notice that the tops of my c-chanel are not in the same plane. How critical is this? Did spreading the legs take care of it? Also my X-Rail is almost perfectly level and straight over the course of my six foot level, there are some places where you can see light (the thicknes of a sheet of paper or less) these areas are not over the bolts, so I dont think shiming will help. The gap is so small I can't tell if its the e-rail or the level that I use. Any coments

04-08-2005, 09:08 AM
I have just attached my x-rails, and have some questions.
I spent eons getting the rails level, finally ending up better than 0.1mm (0.004"), better than 0.05mm over 85% of the rail. Measurement was done with a feeler guage between straight edge and rail, moving the straight edge many times to check (also reversing and flipping the straight edge, in case it wasnt straight!)
Straightening the rail one side was ok, again with 0.1mm feeler guage gap in a couple of places.
So my first question. Have i acieved what I set out to do, or do I have to go back and get it better?
Then I lifted on the carriage and ran it up and down to straighten the second rail to the first.
After some fiddling I arrived at this situation.
3 of the 4 wheels run smoothly on the V, with no signs of lifting, even using a magnifying glass. The fourth wheel, the one closest to 0,0 on the table consistenly runs the entire length on the inner wheel, as though the rails were too widely spaced. As far as I can judge, about 0.8mm.
At the 0 end of the table I think the nearside x-rail needs slight adjustment at the last bolt, but because its off along the whole length, I don't know how to cure the problem.
Any help would be gratefully listened to.

04-08-2005, 09:37 AM
Hi Mike,

One thing to check...

the distance between the V rollers across from each other, my x car got slightly compressed in transit and one corner got moved in by a 1.0 mm The flange the v bearings are mounted on had a definite bend to it on one end.

Flip the X carriage on its back and measure with a metal rod and vernier or a really good tape measure between bearings

If it is out you can smack it back into shape ( what I did) or you can use the shim washers provided in the kit ...which I found later.

Cheers DvanR

04-08-2005, 09:44 AM
Hey Dick,
Thanks a lot. If you don't mind I'll try the shims first.
I'm not sure how hard an Aussie 'smack' is. Somewhere between the Sydney pink parade faithfull and an Aussie Rules forward I guess

04-08-2005, 11:00 AM
Your care and attention with feeler gauge, flipping straight-edge and magnifying glass sounds very good. Don't try and move the rails to match a wonky gantry - trust your straight-edge first.

Shimming will be the good way to get your wheels equidistant from each other. Unless something obviously bent during shipping, in which case you need to to preserve evidence for insurance claims etc.

Shimming should theoretically affect the squareness of the gantry to the x-rails, but you can cross that bridge when you measure squareness.

04-08-2005, 11:43 AM
When you said in another thread (which I have now moved from, as the table is fantastic
) to use a feeler guage, you ommited to say what thickness. I abandondoned the 0.05mm after about an hour and half, and went to 0.1mm. Is this a tight enough tollerance? The magnifying glass is because, as you always say, I'm very old, and can't trust measuring without it! I read a lot of old posts last night about y carriage hold downs. If I roll the y carriage forward and back it get s very 'tight' over the last 20cm at each end. Any thoughts why? (I haven't progressed to 'y' carriage in the manual yet, still trying to sort out 'x'.
One last thing, my manual is an Alpha manual, and the connecting up of the control box bears no relevance to me. Am I going to face problems connecting up? It looks like just plugging together to me.

When do I get the stressful bits out of the way and get on to the fun?


04-08-2005, 12:44 PM
Is 0.1 tight enough....I already think that you are doing quite well enough - for the first time around. There is no harm in leaving it at 0.1 and coming back a few months later to tweak again. As you will have noticed, there is flex in the gantry, and the wheels will find and "average" position (within limits)

Why does the y-carriage get tight to the ends......that should be answered by SB because they set it up. Logically, your 4 rolling edges cannot be all parallel to each other, and 8 V-rollers are trying to follow different paths, fighting with each other. If that was my machine, I would take out the so-called "hold-down" rollers - we still don't use them, even now with the 1/2" Makita router.

Plugging together the pre-Alpha....yeah that was pretty easy. We actually put the motors and controller on the kitchen table and got them to turn before powering gantries and cars. That gave us a lot of confidence.

When does the fun start......this is the fun part - havn't you noticed it yet?

04-08-2005, 01:10 PM
Typical engineer.
Not happy unless he's up to his elbows in grease, surrounded by cogs and ballbearing, with his 4 decimal place micrometer by his side.
Just the sight of a spanner sets me trembling

By the way, send over the dunces cap. Only yesterday did I realise the need to make a bracket to hold my existing (and just overhauled) Makita router. 1200W should be big enough for my immediate needs I hope.
I guess I won't be cutting 'til mid week


04-08-2005, 01:58 PM

At least you didn't pump uk 3 phase into usa 3 phase columbo vfd controller and vacuum pump controller.

The result was a very loud bang a lot of smoke and I'm left with the biggest paper weight the uk has ever seen.

To say I am very upset is an understatment...

As I'm not very technical I asked over the telephone what was needed for a uk spec machine with the info recieved from shopbot ordered my machine.

Brady Watson
04-08-2005, 02:02 PM
No actually Mike, a 'typical' engineer, not indicative of a shopbot owner, will design something that cannot easily be done ~ with no regard for productivity and efficiency. Very few engineers these days actually get dirty...I've worked in manufacturing my entire life and it is absolutely amazing how wide the gap is between the typical engineer's mind and the real world.

Take a look at your engine compartment on your car. It seems that the design team's mantra is, "Nope...too easy to get the oil filter off. Why don't you go ahead and move it closer to the steering rack."

We had this cartoon hanging up in the production manager's office:



04-08-2005, 02:05 PM
I am somewhat nervous until everything is up and running.
The control box says, european 230v, and is set at 230, so I guess that parts ok.
As the stepper power comes via the control box, I guess thats ok also.
Running an existing router, so thats ok.
What was ShopBots solution for your problem?


04-08-2005, 02:35 PM
Hi Mike,

The main control box is fine as it was wired for UK 3 Phase.

Although I only ordered what I was told by shopbot I would require, I've been told that I have to share responsibilty as it was I who filled out the order form.

Shopbot have agreed to supply the replacement parts at cost although I will have to pay for shipping and duty.

My only concern is cost as I am on a limited budget.

I expected to have a steep learning curve and break a few bits on the way as I am not a technical type of person (i.e. I'm a salesman)but I didn't expect my money to go up in smoke.

Yours a very sad Rob

04-08-2005, 02:47 PM
If you track through this forum, and saw the emails I an getting about the European event in September, you will see two things, an ever growing interest in the ShopBot outside of America, and a nervousness of non-Americans in how there purchase will go.
It's a pity that this is not 100% yet recognised by ShopBot, and they aren't a little more understanding when problems like yours crop up.

Write personally to Ted and explain your position.


04-08-2005, 03:23 PM
Robert, what liability is your "licensed electrician" carrying in this debacle? (I see that SB says you must employ one to install a spindle). If SB had a policy of a test run on the destination country's supply before despatch, then the smoke would be made in the right locale.

Mike, how are you going to get this thread back to its topic?

04-08-2005, 04:21 PM
Dear Mike and Gerald,

We used licensed electricians but I assured them that the equipment was for UK use because that is what I believed I had ordered. They wired up the 3 phase wires to the 3 phase wires.

At least nobody was hurt. It really was very frightening and I have never seen my employees move so fast.

The next problem I have is that the spindle is wired for USA use and is not UK use. Shopbot have told me that the spindle can be rewired. I don't know who to turn to who is capable of rewiring the spindle and will this mean my spindle warranty is invalid?

04-08-2005, 04:22 PM
I am waiting for the shopbot salesman to contact me with costs of the replacements equipment and I assume that Ted is aware.

I'd like to thank you both for taking the time to talk to me, this really is a great support forum.

Yours Rob

04-08-2005, 05:53 PM
Hi Robert,

Hopefully you have gotten the email note from me. I was distressed to hear about the problems with getting the incorrect VFD and, as I indicated in my note, we are working to get it taken care of in a reasonable way. I'm sorry for the difficulties and hope to get you running as quickly as possible. Thank you for your patience.

For others who have followed this thread, I can only reflect that we live in a pretty complex world (I'm speaking electrically, or course), and that the specs for this one slipped by our checks. As you might imagine, we are already working on a new data sheet of voltage/phase information for all the countries we ship to, information that will help us double check all orders for appropriate selection of electrical components for each specific locale.


04-08-2005, 11:49 PM

Thank you very very much for your email.

I would like to say that the service you are providing in this matter is fantastic.

I wish that many of the UK companies I deal with had your level of customer service which I can only praise.

For the info of anyone reading these posts, Shopbot have informed me that they are to replace any damaged components and reimburse me any costs incurred.
Should anyone ever have any doubts about Shopbots commitment to its customers I suggest they contact me.

Shopbot have acted in a most reasonable way.

Once again Ted thank you for your help in this matter.

04-09-2005, 01:29 AM
Glad you 'squared away' your problems, now I have to get back to 'squaring my table'

Almost back on topic!

04-09-2005, 02:37 AM
Ted, it looks like you personally have to run the export support desk. That is a square desk isn't it?

04-09-2005, 06:52 AM
Much better day

Fixed the 'off' wheel problem with an "Aussie kiss", see here (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=7&post=22979#POST22979).
My manual is an Alpha manual (I have the last of the PRT96's, which is a mongrel, Alpha everything apart from motors and control box) and it said the motors bolted through the hole farthest from the spring adjuster hardware. No way could I make that fit, so I moved them to the hole closest to the spring hardware, and they fit fine. Tightening the 'one-chance-only' grub screws in the pinion pulled the pinion inwards toward the motor each time, by about 1mm. I allowed for this on the y motor. So now my pinion is offset from the rack by 1mm! One thing suprised me, the force needed to move the carriages once the motors are in place. Not a fingertip job. A remaining question from this mornings work (had to come home early to watch the Royal Wedding, although I think there's going to be one of these a month in future) is there any way to guage how tight the motor spring adjusters should be?

It's beginning to look like a ShopBot

04-09-2005, 09:04 AM
That resistance you feel in the motors is what keeps the SB square while it is switched off. With the motors off, you can coerce the gantry into a square position and it will stay like that forever - unless you bump it.

Don't worry too much about the "once only" grubscrews - better to get them right and tight. link (http://www.talkshopbot.com/forum/messages/312/1185.html#POST6946).

The spring tension thing has always been vague to me, and the current springs may be different to ours, so I'd rather not say anything on that one.

Windsor is a suburb of Slough ain't it?

04-09-2005, 09:23 AM
Now at last I understand what bumping into the stops to make square means. If I can be serious for one moment (about as long as I can be) there are some things you simply can not understand until you have your hands on the actual machine. When I was setting the second stop, with the carriage touching the first, I had a little play if I pulled hard on the carriage, I see now that once set, it doesn't matter.
If you could try the link again, it didn't work. I guess your telling me I can undo and re-tighten them.
The spring tension is something I guess I'm going to have to try to work out for myself.

I should be careful suggesting detrimental things about Windsor, she can still send people to the Tower, you know


04-09-2005, 09:28 AM
One other thing, the manual suggest using a torpedo level (whatever that is!) to set the Z axis square. But this would need a 'bubble' level table to work. Elsewhere on this forum there is much said about this levelling being unnecessary, and a slight slope is no problem.
Can any elucidate?


04-09-2005, 09:33 AM
Mike, refresh the page and the link should work now. We tighten our springs till the coils move apart.

(I have actually twice used the hotels in Slough that overlook the motorway and Windsor park - they have good weekend rates. Plus Heathrow, Eton and Windsor are close by. That's before my father-in-law moved out there. They live in Bracknell now)

04-09-2005, 09:35 AM

Quote *One thing suprised me, the force needed to move the carriage once the motors are in place. Not a fingertip job* ....Be very careful about moving the gantry with the motors connected to the control box or else it will let the smoke fairys out.........Do you know why they built Windsor Castles under the flight path of Heathrow airport??


04-09-2005, 09:47 AM
Torpedo level (http://images.google.com/images?sourceid=navclient&ie=UTF-8&rls=GGLD%2CGGLD:2005-07%2CGGLD:en&q=torpedo%2Blevel). If you havn't got a router yet, then don't bother squaring the z. You need to work from a "surfaced" table to square the z.

Using "bubble" methods, the steps are:
1. surface the table with the router
2. level the table by packing shims under the legs
3. square the z with a torpedo level held against the z-slide rails
4. square the router spindle to the table with a trammel
5. re-surface the table with the router.

Using a framing square (http://images.google.com/images?hl=en&lr=&rls=GGLD%2CGGLD%3A2005-07%2CGGLD%3Aen&q=framing%2Bsquare&btnG=Search):
1. surface the table with the router
2. square the z with a framing square held against the z-slide rails
3. square the router spindle to the table with a trammel
4. re-surface the table with the router.

04-09-2005, 02:15 PM

In answer to your question, was it because Her Majesty wanted the triple glazing grant?

Got the router, haven't got the bracket, I'm stealing the design from some photos I found here (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=2&post=15264#POST15264)

Can you expand on "3. square the router spindle to the table with a trammel" please? Do you mean for me to use some type of right angled attachment fitted in the collet? Turn it through 360 degrees?

Can I do some air cuts first? (Please,I want some fun)


04-09-2005, 02:25 PM
For fun, tape on a pencil and write your name.

As for that bracket design, I even saw it being used on another SB in Cape Town. Do you think that I should mass-produce it for sale?

Do you mean that you've never seen trammels mentioned here before? Search on the word

04-09-2005, 02:43 PM
You might licence the design

Read the threads, I can now trammel with the best of them.

This from an internet dictionary
trammel \TRAM-uhl, noun:
1. A kind of net for catching birds, fish, etc.
2. A kind of shackle used for making a horse amble.
3. Something that impedes activity, progress, or freedom, as a net or shackle.
4. An iron hook of various forms and sizes, used for handing kettles and other vessels over the fire.
5. An instrument for drawing ellipses.
6. An instrument for aligning or adjusting parts of a machine.

Our needs come 6th


04-09-2005, 03:08 PM
Position 5c at the standard American dictionary - with their pronounciation. link (http://www.m-w.com/cgi-bin/dictionary?book=Dictionary&va=TRAMMEL)

What on-line dictionary do you use? I can't find a free Oxford Queen Camilla's English version. Here (http://www.seslisozluk.com/search/trammel) is a Turkish site with a pretty comprehensive result on trammel

What do you think I could charge for training on that bracket?

Oh, you can either grind the heights of the 4 pedestals on the bracket (equal height above surfaced table top) or shim behind the bracket, to get it square to the table. There.....thread back on track.

04-09-2005, 04:11 PM
A bit late Mike, but I seem to have an old assembly manual in pdf form if reqd........

Sorry to hear Roberts tale of woe. But glad your both ahead of me to sort the problems out.


04-09-2005, 08:55 PM
To square my router to the table I chuck this into my router. It is made from a piece of 1/2" rod and an indicater bolted to the end.


04-09-2005, 11:01 PM
Great jig! Do you use it 10,000 or 15,000 RPM...

Really, I have a similar unit and it works best with a fresh sheet of MDF(smooth).

04-09-2005, 11:18 PM
Dirks device or similar ones found in catalogs such as MSC are the best for dialing in the Z axis and getting one square to the table. The big problem is finding a flat plane to adjust the Z to. It is best to flatten the table surface as best you can first. This will most likely leave ridges across the table surface, unless you are really lucky or darn good. They can be dealt with by placing a piece of plate glass on top of the ridges that is large enough for the swing of your trammel. This will approximate a flat plane as closely as possible. Once you have this flat plane you can chuck the dial indicator trammel into the collet and swing it around adjusting the Z until you have an equal reading all the way around. Now, you're ready to take the trammel out and re-surface the table to get a truely flat table surface that is square to the spindle/router.

If you don't have a trammel, you can use a piece of scrap material and suface a pocketed rectangle into it. Check for ridges and adjust the Z towards the high side of the ridges in X & Y as necessary. Then pocket another rectangle a little bit lower and the check again. Re-square the Z if neccesary and repeat until you have a flat surface. This method, though a bit crueder than using a dial indicator, will produce a flat surface. Note that the larger the diameter the cutter you use the flatter you will get the table this way. Smaller cutters may not produce ridges that indicate the Z is out of square.

Hope this helps.

04-10-2005, 12:32 AM
I use a scrap of wood (roughly 3/4" x 3/4" x 20" long) with a 3/8" drill bit stuck in a 3/8" hole in one end, which I can chuck in the router's 3/8" collet, and any smaller drill bit stuck in a hole in the other end. The small bit hangs down lower than the end chucked in the router. Looks a bit jury-rigged, but it works. By swinging this puppy in a 40" diameter circle, I only need for the large end to be accurate to 1/16" or so. Since my planing bit is "only" 1.25" diameter, that equates to less than .002 across the bit diameter. My widebelt sander will remove .005 even with the finest grit, so that's plenty flat enough for me. The ridges left on the table are too small to feel with my fingertips.

here's the proof (http://www.talkshopbot.com/forum/messages/312/7305.html?1113107331)

04-10-2005, 01:08 AM
would being taken out to lunch be included in the training? Rand or Euros?

It has only just occured to me that you have to get the Z axis square and the router square, independently. I find no mention of this in the manual.

If I keep stating the obvious (as far as experienced ShopBotters are concerned), I'm sorry, but when I don't understand something, or the light suddenly goes on and I realise something, I feel that there may be a few other newbees in the same situation.

A copy of the 'old' manual may be a godsend!Got your email,thanks

I like the plate glass idea. Is a large mirror plate glass?

Thanks as always


04-10-2005, 02:00 AM
"It has only just occured to me that you have to get the Z axis square and the router square, independently. I find no mention of this in the manual." A very important point, and very overlooked on this Forum too. Notice in my job lists above that I particularly said to check the z-axis rails before checking the router with a trammel. (I think many SB'ers have a false sense of squareness because of the detailled focus on the trammel methods).

Note that V-rollers running on the z-rails have eccentric bushes inside to adjust the tension against the rail. After some of this eccentric fiddling, the sliding section is no longer parallel to the tower-frame.


04-10-2005, 02:06 AM
How I square my Y axis:

04-10-2005, 03:30 AM
I have just spent some more time reading the manual, and I can not find anywhere this vital information about independantly squaring the Z axis and router. The Forum only lightly brushes up against it also.
I have found posts which describes correcting squareness where 'ridging' has occured, but I can find none suggest correcting both Z axis and Router (although if I search for longer, they may be there).
An omission that need putting right IMHO.

04-10-2005, 05:01 AM
Mike, I think this (http://www.talkshopbot.com/forum/cgi-bin/discus/show.cgi?tpc=7&post=7567#POST7567) was one of the few references to it here. Everyone is having so much fun devising trammels that I am hesitant to confuse the issue with eccentric rollers and all.

IMHO, a dial gauge on a trammel is an over-kill, plus it takes a lot of walking (to read it from the other side) and remembering little numbers to get readings at the 4 points of the compass. (I tend to do what David does - just swing a long pointer). I had a dial gauge on a trammel once and this gave me different readings depending what part of the table was being used. I saw that the y-gantry actually had a twist in it and a square router spindle at y=10" was not square at y=38". And a mug of coffee on the gantry can even make a difference.

The major parameter to keep square, is the x to y axes squareness - that is the most difficult to measure, but the easiest to fix. None of us use dial-gauge accurate methods for getting these two long axes square, so why do we become so critical when we set up the shortest axis (Z) of the SB? Particularly if we realise that the plunging action is set up only with a torpedo level or framing square?

Who has ever bothered with a dial gauge trammel for a drill press - try that for a good laugh.

04-10-2005, 05:24 AM
This is how your z-slide can be moving in your z-tower - those rollers are all eccentricaly adjustable:

The trammel method will NOT detect this!

04-10-2005, 07:45 AM

I usually agree with what you say but this time I have to totally disagree with you. Your picture showing how the eccentric action can apparently miss align the Z-axis seems to indicate this would be a problem, when in fact it is not. Your statement that "The trammel method will NOT detect this!" is irrelevant, there is no need to. You can have the rollers adjusted exactly as you show but when you square the router to the table you are squaring a parallel line with the face of the rollers. This is your main objective.

04-10-2005, 07:47 AM
That was the only reference I found!
One thing I have had difficulty whilst setting up (to this point) is deciding if I'm close enough.

This is not to scale, but shows the position of the V-wheel on the rail. I have pushed it up and down many times, with two of us observing first front wheels, then back wheels, then swapping sides and doing it again. It never gets close to the right hand drawing, not even half way from what we can judge. But there is that hint of movement in a few small places. Am I close enough?
We listen to the clonk of the wheels hitting the blocks many times, only ever hear and feel a single clonk.

Its all very worrying for a simple guy!!


04-10-2005, 09:52 AM
Frank, imagine cutting through a 2" thick piece of hardwood with a 3/8" cutter, taking 4 cuts each 1/2" deep. With the situation I sketched, your last cut will not be vertically below the first cut. The "water-line" ridges that you expect to find right around the part will not be right around - they will lie to one (left) side only. (as per sketch) In fact, the right side will be trying to undercut each previous cut and the tool might even burn on the last step if you con't have a full 2" cutting face on the tool. But the trammel will tell you that everything is 100%. Think about it Frank......

Mike, those rails are soft and they will "mold" to fit the wheels. Take a break from shimming now and get your router mounted. Then let the whole machine "run-in" and then you can go back to tweaking. Relax! (I've lost your phone number)

04-10-2005, 10:16 AM

You are correct in your illistration and explination as to how it will cut if the tool where adjusted as shown in your picture. But the explination "The trammel method will NOT detect this!" Is WRONG! It most certainly will, not only detect it, but will compansate for it. I have to leave a bit, I will send an illistration with a better explination later.

04-10-2005, 10:27 AM
The trammel method can be executed without moving the z-slide at all. Therefore the trammel is not at all sensitive to the plunge direction - it cannot be used to square up the plunge direction. Looking forward to a good argument

04-10-2005, 10:37 AM
Gerald is absolutely correct in his reasoning. I suspect the reason this hasn't come up before is that it doesn't negatively affect the final cutting for 99.5% of what people are actually doing with their machines. How many times have we covered offsetting the multiple rough passes and following up with a light clean up pass at full depth - the clean up pass removes any evidence of the Z axis not being square to the table. Even with 3-D work, I suspect the fact that the final product is "skewed" slightly in the Z axis is undetectable in most if not all cases.

Squaring (and keeping square) the Y axis to the X axis is, as Gerald, said, more important for most users' real life needs.

04-10-2005, 11:00 AM
Without a doubt trammeling only solves the problem for 1 specific z height. When Gerald pointed this out (and it sunk in) I also realised that what you have now said in your post is so true. I'm thinking that a lot of people have this problem to some degree, but think its minor run out or something similar, and correct with a finishing pass. I have looked back at some of these problems, and the independence of Z axis to router has never been mentioned.
Of course, many are squaring the Z axis well, and correcting errors found in trammeling by adjusting the router, therefore doing the right thing. Its just never been discussed in depth before.
Sorry Gerald, there is no argument, you are right


04-10-2005, 11:38 AM
Jeez, and there I was looking forward to a good argument.

Anyway, here is a sketch, looking in the x-direction, that shows the difference in cut result between the two different issues: Left is effect of a skew plunge cutting 1/2" at a time through 2" thick. Right is a skew spindle, but with a perpendicular plunge:

I prefer the result on the right, and hence I don't bother too much with dial gauges for trammels.

04-10-2005, 12:00 PM
Your not geting off that easy Gerald.

Either we are all talking about something different or you are all not understanding my point.

This picture illustrates how the mounting surface (mounting surface refers to the point where the v-rollers attach too) can be misaligned with the table, as long as there is enough room for the eccentric to compensate for and align the v-roller EDGE 90 degrees to the table. When you make the adjustment with the bolts holding this bracket you are actually aligning the EDGE of the v-rollers with the table irrespective of eccentric position. There is one possible way that Gerald’s statement makes sense and that is if the router axis where miss aligned/canted with the rollers edge. This problem would need to corrected through aligning the router axis parallel to the EDGE of the V-rollers. But my point is that the concentric rollers have nothing to do with this.

The up and down path and the axis of the router bit can be adjusted using the trammel method under this condition. Granted you would not want it out this far as the surfaces at the adjusting bolts would never stay tight since they are grossly misaligned in this illustration. When you adjust using a trammel you are aligning the axis of the router (assuming the router axis is parallel to the V-roller edge) perpendicular to the table irrespective of the mounting surface the V-rollers are attached to.

For Illustration purposes I have exaggerated the angle beyond normal.


04-10-2005, 12:23 PM
Frank, you put a very key statement in brackets (assuming the router axis is parallel to the V-roller edge). I agree with you if we are sure that they are parallel, and in this case the trammel will show if both are perpendicular to the table.

But, I am arguing that we have no way of knowing that these two things are parallel. Indeed, we must assume that are not parallel before we start our adjustments.

It is only after we have have squared the two systems to the table that we can start to assume that they have become parallel to each other.

04-10-2005, 01:21 PM
Geralds picture shows the mounting surface square with the table,the V rollers skewd, Franks picture shows the opposite, mounting surface skewed, V rollers squared.
That is, Geralds alignment needs correcting, Franks alignment has already been squared up(He's done step one).
Now Frank then 'squares' the router by trammeling and adjusting the router.(step two).
However the assembly manual tells you how to align the Z mounting surfaceas per Geralds picture and ignores completely any mention of ecentric rollers. So, having squared the mounting surface how do you adjust the V - roller edge? There is no mention of accepting Franks skewed mounting surface, and correcting it.

(and I thought tomorrow I was going to have an easy day


04-10-2005, 01:46 PM
Relax Mike! Go back to the task lists above and read carefully "...held against the z-slide rails". In other words, hold your squaring instrument of choice against the V-ground edges of the slide. You can even hold the instrument against the rollers, seeing that they are equal diameter. Don't start fiddling with the eccentrics.

Could you mail me your phone numbers again?

04-10-2005, 01:54 PM
That's correct the eccentric’s main propose is to make the Z carriage track without free play as much as possible. There porpose was not to make alignment adjustments.

04-10-2005, 02:16 PM
Gerald, I don't believe we are disagreeing with what is necessary to obtain a well-adjusted machine. It was just that your statement that the” rollers are all eccentrically adjusted” then looking at your picture it appears you are conveying that this would be a problem. When in fact you could have a Z mounting bracket and carriage set up exactly the way your picture illustrates and still adjust the mounting bracket using a trammel. That is assuming your routers axis is parallel to the edge of the Z mounting plate. If your picture had shown the routers relation to the Z mounting plate I would totally agree you would not be able to get a good adjustment.

04-10-2005, 02:30 PM
When you say,"That is assuming your routers axis is parallel to the age of the Z mounting plate", it's clear you understand the overall problem.
It was understanding this necessity which wasn't clear to me at the start, and which, because there's nothing in the manual about it and, before this thread, little to nothing on the Forum about it that makes this discussion rather important.
It's interesting that, even when people are talking at cross purposes, interesting and useful information can be brought out

04-10-2005, 02:32 PM
Mike, I may have read but forgot. Are you using a spindle or router? The reason I ask is that it would be more difficult to ascertain whether the router axis was parallel since I believe it is an L bracket. The spindle is a different matter. It is bolted directly to the Z plate and very easy to ascertain whether it is parallel or not.

04-10-2005, 02:48 PM
Frank, I still disagree with you

Before you fit a router to the ShopBot, you can square the z-slide to the table - where will you fit the trammel to do this squaring?

04-10-2005, 02:58 PM
I'm mainly cutting 4mm and 6mm Birch ply, occasionaly 15mm softwood, and not in mass production numbers, so I'm hoping the router will do the trick.
I am going to copy Geralds mounting bracket, at a licence fee though


04-10-2005, 03:03 PM
After that last mischievous question I must say that I see where Frank is coming from, when he is using a chunky rectangular spindle and I am using a lowly router on a home-made bracket. Frank is reasonably confident that his spindle is parallel to the slide when he starts off, so the trammel is enough for him. But us guys using funny-shaped Makita's and PC's, sometimes with hoseclamps, cannot assume that we have the router axis parallel to the slide axis, and thus have to go through two separate steps.

04-10-2005, 03:06 PM
Gerald, I do not dissagree with you. There is no need for a trammel at that stage of alignment. A large framing square does an excellant job squaring the Z slide to the table.

Brady Watson
04-10-2005, 05:36 PM
Lot's of mountain climbing over molehills here I see...when a level would work as the manual describes.

I have achieved excellent results aligning both Z-axes on my bot, and other PRTs using a good quality 2' level clamped to the 2X2 uprights on the each Z. I specify a 2' level over a small torpedo, because you can read the top bubble and get a more accurate reading. If you followed the instructions correctly from square 1, you will see the importance of leveling your table...since whatever spoilboard is on the table at the moment hasn't been surfaced, ALL of your methods of using it as a reference point are unreliable. The manual recommends a level as the correct tool for a reason.

If you are super particular, take a digital caliper and use the depth gauge to read the distance from the 2X2 to the ground angle on the Z-axis. If it is out, which I seriously doubt that it is, then make adjustments to it 1st. Then bolt up the Z axis and level it both ways, checking all around to make sure you are still level. Viola! DONE! If you have done this correctly, then when you surface your spoilboard, it will be both level and perpendicular to the Z-axis and there should be no ridges in the table either.

Leveling the Z correctly is a 10 minute job at best...no sense in 'arguing' for 4 days about it. Even if your machine was perfect, expect it to be out .008 from the start with a standard router.

Mike: Get it as close as you can without going nuts...and start learning to use the machine. If it is out, you will know it. And...you will get experience tuning it. Plan on killing that spoilboard and replacing it in the next few months, since you are a new user...it's gonna get 'spoiled'


04-10-2005, 10:05 PM
If you are creating a piece that is for arguments sake 4' by 4' and your axis in any direction is our .0001 there is in the world of real woodworking whether you are carving carousel horses or cabinet boxes no difference what so ever. All tolerance numbers exist as guidelines for comparisons sake. Put the darn thing together. Cut some parts. Move on. It doesn't really matter how many angels can dance on the head of a pin.

04-11-2005, 01:11 AM
Brady, you said "...... good quality 2' level clamped to the 2X2 uprights on the each Z.". Are you missing the point that the z-slide direction is not necessarily parallel to the 2x2 uprights? (that sketch of mine above is not a huge exaggeration).

04-11-2005, 02:13 AM
The problem for me is the many posts in this forum that talk of exremely fine tweaking, and for me to decide when Im close enough.

The manual does instruct on how to align the Z axis using a framer's square. As both Frank and Gerald point out, its the V roller guides that need to be straight, not the uprights. I had to hit a part of the x-carriage with a rubber hammer to correct the 1mm (0.04") error on the seperation of V-wheels. If there is a similar error on the z carriage, aligning the struts does no good at all.
However I am going to put it together, run it, and see what I get


04-11-2005, 02:31 AM
Mike ,

Look at the trammel you started.....

Your original question was to do with the v bearings. Need to be sure the picture to the "right" is not the middle one ( which I think is right rather than wrong) and not the far right one which I think you are trying to achieve. ( and is wrong)


04-11-2005, 03:43 AM
Have just measured the misalignment between our z-slide and the z-tower frame (2x2 square tubes). It is about 1.8 to 2mm over the height of the tower (about 500mm).

It could well be that the SB folk align the slide to the tower before shipping, but ours arrived with one of the four wheels jammed under the rail. With the fixing of that problem, and with occasional eccentric adjustment to compensate for rail wear, we are now at a situation where there is quite a large deviation between the line of the slide and the tower frame.

04-11-2005, 03:51 AM
Just how do you align the slide to the tower?

04-11-2005, 04:04 AM
To align the slide to the tower, you would have to fiddle with the eccentric bushes in the V-rollers. The eccenters allow an adjustment of +/- 0.61mm (1.2mm or 0.05" total). In the early days we were nervous of adjusting the eccenters so we ended up with the tower sitting slightly skew in the y-car. Brady would measure our tower and tell us that our SB is way out of square - but it isn't

04-11-2005, 04:20 AM

You state “In the early days we were nervous of adjusting the eclectics so we ended up with the tower sitting slightly skew in the y-car” so if you agree that the tower can be slightly askew and still obtain a good adjustment then why would you want Shopbot to try to align it using the eccentrics.

04-11-2005, 04:35 AM
For the guys who hold the spirit level against the tower tubes - (apparently as per the manual).

04-11-2005, 05:06 AM
Okay, I pulled out the manual for the older PRT it does show placing a level on the upright square tube. This should certainly be sufficient for a preliminary adjustment. Remember at this point you only need to get close, since you will not be making the final adjustment until you have mounted your router and installed your spoil board.

I did not use this method instead I slid the Z mounting plate down as far as it would go and placed a framing square on its edge.

04-11-2005, 05:37 AM
Your earlier remarks about carousel horses (actually Rocking Horses in my case) is spot on. I leave about 1/4" (6mm) of wood to be taken off be hand, to make a truly hand finished piece, so water marks and sub 1mm errors I can ignore. The mechanics of my automata are also so small that these minute variations will not effect me.

Why hasn't Trammel been added to the 'new' words thread?
The middle picture is what I tried to achieve, the right picture is wrong

I have just squared my Z axis the same way you did, slide it down and measure the V edge with a square. I did notice as I tightened the 4 bolts one by one I had to keep giving little taps to keep it spot on.
The Alpha manual also shows the same picture of the spirit level against the tube.

I am at home having lunch at this time, but when I return tonight with the manual I will start a thread with my perceived manual errors. Then you can all enjoy shooting me down and putting me right

Now to plug it in !!!!


04-11-2005, 05:56 AM
Yes, the method of adjustment is a bit crude but in the end it doesn't work so bad. In the 80s and early 90s when I was installing satellite antennas many manufacturers used similar methods of adjustment. I would usually rig a turn buckle to make the adjustment easier. Later on many manufacturers added the buckle but some never did. It was a bit tricky making adjustments on a 12 foot dish when a movement of only 1/4" at the age made a difference between being on the satellite or no picture at all. The dish pivoted on one single three quarter-inch bolt.

Brady Watson
04-11-2005, 07:54 AM
No Gerald, I get what you are saying. If you are concerned that the slide is out, attach the level to the router itself. There....problem solved. You guys are a trip! Just put the thing together and run it already! The proof is in the pudding...start making some parts!


04-11-2005, 08:05 AM
The right thing for attaching to the router is a ...........trammel. And then we can can go full circle and start this "argument" from the beginning again.

Brady, some of us spend weekends away from the shop, so we procrastinate at the Forum instead.

04-11-2005, 08:19 AM
Seems to me there's no argument to start again, so just get out the trammel and get it done.

04-11-2005, 08:42 AM
There you go Mike....Even if your setup looks like this to the eye, but the router is trammeled and plumb level to the table, it will apparently work fine. That is what they say.


04-11-2005, 10:06 AM
The bit would be straight but the y position would change as you go up

04-11-2005, 10:38 AM
So there I was with the straightest z axis this side of the Rockies, all my wiring nicely tied away, computer and control box humming merrily, making contact, when the input number 4 light flashed to tell me the emergency button was 'on'. I'd forgotten to connect it! Looked at the end of the cable, 4 wires, no plug. Looked at the manual (and 1 sheet supplement); instructions only for the Alpha, nothing for my poor old PRT96 Hybrid
Gentlemen I ask you, do I deserve this amount of dissapointment at such an advanced age?
I think I have to wire in the red and white wires, but to which terminal?
I will now search the Forum for the answer!


04-11-2005, 11:15 AM
Red to Ground, white to #4 ?

04-11-2005, 12:17 PM
Mike, Thanks for all the questions and clarifications. I'm learning things I should have known long ago. I have been trying to email on Eurocamp and other subjects and always bounce. When you have time, drop me a test message and I'll try a reply.

And for the rest of you, thanks for helping Mike as well as many unseen past and present newbies better understand and use our machines.


04-11-2005, 12:33 PM
(and anyone else trying to email me)
My apologies. My web server is being changed so I've amended my email in my profile to miketeejay@hotmail.com (mailto:miketeejay@hotmail.com)
I thank you for the sentiments.

04-11-2005, 12:42 PM
Hi Normand, I agree with you. Your website, and your work is very impressive. I like the vice/clamp that holds your work in your profile picture.

Mike, entirely at your risk, put a paperclip from ground to #4 directly at the board - you can figure out the switch wire colours later. (VD,,1 for mm)

04-11-2005, 02:38 PM
Following Robert's disaster, and with wiring day getting closer, I had a look at my control box. I ordered 240v single phase, and this is what the invoice says, but the input wire has 4 conductors - black, white and red, plus green. I'm no "licensed electrician" (although I do have an electrician qualification), but I expected 3 conductors: +, - and ground. That's all that's available where its going to be connected. Can anybody tell me what's going on here?

I can't see a plate on the control box specifying what phase and voltage the box is, although stuck to the box was a 'post it' note with "Hesp John 230V Indexer" written on.

Mike, I emailed the old manual to you yesterday. No use?


04-11-2005, 02:55 PM
The manual has already helped, thanks!
My box was wired OK, but it is, of course, an old PRT96 control box, so that doesn't help you a lot.


04-11-2005, 03:39 PM
I assume you have consulted with Shopbot directly on this? Better to have them avoid the smoke than to be discussing who should pay for repairs.


04-11-2005, 04:19 PM
Hi John,

Your Control Box is set up for UK standard 240v single phase (& single pole in our terms). It uses 2, 240v circuits, one for the running your ShopBot, one for the router (we're assuming here a 240v single phase router, not a spindle).

The RED WIRE to the Control Box connects to the LINE for the first circuit, the BLACK WIRE connects to LINE for the second circuit. White is the common NEUTRAL and GREEN is Ground.

You can confirm how the wiring works by following from where the wires leave the main disconnect in the Control Box. Red goes through one set of contacters to run the router, Black goes through the other set of contacters to run the drivers and Control Box power supply. White goes through the contacters to both circuits. Green goes to the ground bar.

Sorry if this seems a little confusing. We have something like 12 different power arrangements for Control Boxes and it gets us confused at times. We believe that the wire colors and the way they are routed should make sense to your electrician. Please send a quick email to support@shopbottools.com (mailto:support@shopbottools.com) (or call) if this is not clear. We do check emails as frequently as we can.


04-11-2005, 05:16 PM
Thanks Ted, that seems clear enough. Maybe my electrician would have realised this, but I didn't want to be paying him to scratch his head.


04-11-2005, 06:29 PM
Thank Gerald, the vice device is my old washing machine ,a car jack a radialarm saw arm 3 vise and soapstone brick for weight. Take a look at those pre shopbot pat. us 2,793,569 DE 3,112,361

04-12-2005, 01:41 AM
Ted, the European Union is busy "harmonising" their cable colours across their member countries, which should make life a bit easier for you. Here (http://www.niceic.org.uk/downloads/WiringSupp.pdf) is a fairly clear guide (written from a UK perspective) to what is mandatory across the European Union from March 31 next year (2006), and has been optional from March 31 last year. Many other countries outside Europe will probably follow suit with time.

Colour is only one part of the issue - terminology and markings are also very important.

John, don't be alarmed if your electrician connects red and black to each other. If you have a single phase supply, he doesn't have any other choice.

04-12-2005, 05:02 AM
Gerald, I assumed that they would run seperately back to the fuse box, and that they would be connected together inside the control box if connecting them together was an option.

04-12-2005, 05:29 AM
John, if you have single phase, then you only have 3 possible terminals (L,N,E). ShopBot has supplied you with 4 wires to connect to 3 terminals. Ted speaks of "LINE for the first circuit" and "LINE for the second circuit" - in your case you only have one "LINE" available. If you or your electrician are uncertain about this, then I suggest that you ask ShopBot if you could connect their two "LINE's" together onto one pole. I agree with the logic of your sentiment "that they would be connected together inside the control box if connecting them together was an option" but we seem to be worlds apart when it comes to logic.

04-12-2005, 06:56 AM
Gerald, certainly I am uncertain about this, and if the electrician is we'll contact SB.

Although, as you say, I only have one "LINE" to connect the two ShopBot LINES to, I think they might need to be connected to it via different size fuses.


04-12-2005, 07:09 AM
John, independent fuses for the two lines might be a duplication of what is in the control box already - I have heard that there are fuses in the control box. My personal choice would be to run both lines off one external fuse - so that a "tripping" router shuts down all movement (machine damage risk), or that the router stops turning if the movement "trips" (fire risk). But please let SB and the electrician guide you.

04-12-2005, 07:19 AM
To me, the question about "where" to connect the LINE conductors is a question about fuses. Whenever possible, when I have loads drawing high current, I like to have separate breakers (fuses) for each load. Running separate LINE conductors to a breaker panel, rather than tying both LINES together inside the Shopbot control box makes sense to me; however, I am not an electrician; therefore, I have no idea of what is allowable.

04-14-2005, 04:28 PM
Mike, you hit the nail on the head. The logic is to have the router and the drivers protected in the best way. The wires are intend to supply 2 independently protected LINES. Having a breaker or fuse for each is important and helps prevent the wire size from getting out of hand. We could have set it up to use one LINE, but the size of the cable and the fuses would be very large. The cables that supply power to the control box are not sized for the amount of current that would be required from a breaker large enough to handle both loads. In short, each LINE wire needs its own fuse or breaker to protect the wires as much as the components. The extra fuses in the control box are just for that prupose and are not intended to be the main fuses protecting the system.

Please have your electrician size independent fuses for your controls and router or spindle according to your local electrical codes.

Gerald, thanks for the wiring guide. I have studied the EURO standards for wiring and coloring and will use this as a further reference.

You guys have really beat the squaring issue out. You have identified all of the issues I can think of. We have kept in mind the squareness of a router bracket and align them to the T-rail during assembly. This is simply done by squaring the top of the router mounting bracket to the edge of the T-rail. We are confident enough in the machining of both of them to trust this method. This gets a bit more complicated and not easier with a spindle. It is important to make sure that your spindle is parellel to the edge of the T-rail before proceding with squaring the Z to the table. Be careful when adjusting the eccentric bushings, if you get the rack too far away from the pinion you will introduce backlash and possibly have trouble with the pinion disengaging the rack and the Z dropping. Adjust them in moderation!

04-15-2005, 12:38 AM
Hi Gordon, you said "The cables that supply power to the control box are not sized for the amount of current that would be required from a breaker large enough to handle both loads". I hope that the NEUTRAL (white) wire in your cable can carry the sum of the curent in the two LINE wires (red + black). In three (or two) phase systems, the neutral wire can be relatively thin, but in a single phase system the "forward" and "back" currents happen at the same instant. Having understood your logic, I would expect to see that the white core of your cable would be thicker than any of the others?

If we had to provide two "fuses" to power one piece of equipment, and they are the "disconnect" devices for that one cable, then our code requires that the "circuit-breakers" be mechanically connected so that they cannot be disconnected independently. This is probably for the safety of the electrician working on that cable/consumer.

04-15-2005, 01:01 AM
As this thread has drifted ever so slightly
from the original subject, is anyone able to answer this power question for me.
On page 116 of the ShopBot users guide, paragraph 4, it says (refering to electrical noise) It is also best not to place your ShopBot Control Box and computer on a seperate circuit from the router and dust-collector, and other shop equipment.
Circuit to me means a ring including a number of outlets coming from a single fuse (or cut-out) in the main power box.
It seems neither possible nor desirable for me to put all my tools on a single circuit.
I have in-line equipment to take care of power spikes.
Any advice would be appreciated.

04-15-2005, 01:19 AM
Looks like a misprint in the manual? See this post (http://www.talkshopbot.com/forum/messages/26/1319.html#POST7692) by Ted.
(found that link by a search on "separate circuit" - with the little "marks in the search box)

04-15-2005, 01:27 AM
I posed this to Frank at support, he asked for the page number in the manual.
I will report his answer (which will presumably be ShopBot definitive) back here.

04-15-2005, 08:46 AM
Referring to the NEUTRAL (white conductor) in Gerald's post, I would assume that the NEUTRAL conductors are connected to a common 'buss' bar or terminal strip inside the control box (similar to the 'buss' bar inside a breaker box). In that way, the NEUTRAL conductor going to a distributed circuit from within the control box can have the same guage conductor as the LINE conductor for that circuit. Separate cables can then still be used outside the control box, each connected to its own breaker in the breaker box. The load on the NEUTRAL conductor would then be distributed between the loads inside and outside the control box. HOWEVER, if a single four-wire cable is used to connect the control box to the breadker panel (LINE-1, LINE-2, NEUTRAL, and GROUND), then the NEUTRAL conductor would have to be sized to carry the sum of the currents of LINE-1 and LINE-2.

Now referring to having the control box, router, dust collector, etc. all on the same circuit. I would assume that the intention was to have all circuits common to the same breaker box to avoid the possibility of 'ground loops' and to avoid the possibility of having one sub-panel 'trip' while another sub-panel remain on. If the sub-panels were in physically distant parts of the shop, a very dangerous situation could occur. Inside the breaker box that controls all equipment in the shop, the NEUTRAL conductors of all circuits would share a common 'buss' bar, as would the GROUND conductors share a common 'buss' bar.

04-15-2005, 10:31 AM
Further to my post (4 above this one) Frank at ShopBot support was quick in coming back and said this "It appears that there is a NOT that should not be there on page 116. The standard practice has been to isolate the control box and the computer from any possible noise spikes.
The same as Ted said back in 2003.