Confused with 13 ply Bot losing steps position

[jhicks]

Anyone out there cutting 13 ply on a regular basis? I have been running many types of sheet stock, solods, plastic, foam etc but have frequently encountered "stalls" when cutting 13 ply. Ive been cutting with a 5mm bit at anything from 13000 rpm up to 17,000 rpm, and feed from 160IPM down to 120 IPM to try to dial this in. The problem is parts will cut fine for a while 2 to 4 sheets into a brand new bit, then suddenly decide to stall, lose position, and start cutting an entirely new pass in the wrong position. Cutting depth .780 in 3 passes.
Latest sheet had about 18 parts on it. Many cut fine on all 3 depth passes for the cut outs, then suddenly the Y axis stalls, loses position, and I'm off 3/4" in the final pass. ARGH^%#$^$
This has been happening the past couple days and is really confusing me not to mention taking endless hours to recalibrate position and start over.
My speculation has been centered on something to do with the material, like lots of glue, who knows what the core layers are? A hot dog mix of chinese whatever I guess. No visible metal, barbed wire, or nails isolated.
The problem is I'm nervous about cutting the "good stuff" like mahogany, and other cores like mdf, 7 ply, prefinished maple etc when I cant really isolate the problem.
I've had this happen before even with a 3/4" compression on 13 ply so just looking for any feedback from anyone with similar experiences or maybe an answer.
Just seems like the motors can not plow through this stuff and get to skipping for some reason.
Thanks

[bill.young]

Hey Jerry,

It sounds counter-intuitive but can you increase your move speed and try cutting in 2 passes? I know you're using a smaller bit so this may not apply and I can't speak for the 13 ply, but I just finished cutting around 500 sheets of 3/4" 7-ply BC pine on 2 PRS tools. I cut them with 3/8" 2-flute bits from Centurion in 2 passes, with the spindle set at 11,000 rpms and a move speed of 6"/sec. Most of the parts had to interlock like puzzle pieces so cutting accuracy was critical, and after some testing and fiddling these were the settings that worked the best. I averaged around 15-20 sheets per bit, depending on the number of parts in the sheets, before the bits got dull enough that cut quality started falling off

One thing that I finally realized...I'm kind of thick!... was that even though cutting in lots of passes puts less load on the tool, it's only a short section of the bit that's doing all the cutting. So if you're cutting 3/4" in 3 passes it's only the bottom 1/4" of the bit that's doing all the work (and getting all the wear). If you do it in 2 passes then the bottom 3/8" is cutting...a 50% increase.

Bill

[gabepari]

I cut 3/4" Baltic Birch (well 18mm to be exact) a few times a month. I cut at 5 ips, 15K rpm with a 3/8" 2 flute straight cutter (amana carbide tipped). I use my usual path strategy, first cut 80% deep in the climb direction, followed by .010" into the spoiler in the conventional direction. If I require no tearing or chipping on the corners I will switch to a compression cutter, but everything else stays the same.

Works for me.

Oh, if it matters, I have a PRT-Alpha 5x10 with 3hp colombo, and 3:1 belt drive boxes on the x and y axis.

Gabe

[jhicks]

Thanks guys, I understand the 2 vs 3 pass Bill and get the bit wear but for whatever reason (weak motors?) or reasons dont fully understand, when I ramp up the move speed much past 3IPS / 180IPM I experience more stalling regardless of 2 or 3 pass even with a new bit 5mm or 3/8" on my PRT 120 with a 5hp columo spindle.
This is making me think my power cables may have intermitencies. But I've never really been able to run/cut effectively much past 3 IPs or 180 IMP on this machine via sb controller or Ascension and marvel at those who do like Gabe. I can definately move at rapid 425IPM + for rapid jogs but cutting is an entirely different animal for me. In fact way back when i was less knowlegeable, I was told the PRT really wouldnt run much faster by SB support so thats kind of been what I've come to expect. I guess i've assumed it was the motors but its still a mystery to me.
Not sure what my gear box really is (was there a standard in 2003 or how would I find out/confirm?)but I can push against the gantry as its moving at 160IMP and create a missed steps and hear others dont experience that either.
HMMM???

[gabepari]

Yeah, the Alpha motors were definitely a good buy. I routinely cut 3/4" MDF with the same 2 pass strategy at 9 ips. And jog at 20 ips (though I don't think the thing ever sees 20 due to ramping and short jog distances between parts, but hit JH and the thing flies back to 0,0).

I built my own gear boxes, don't know what was standard on your machine.

I know these aren't impressive numbers compared to the "real" machines, but works for me.

Gabe

BTW - check your grounding as well, I had quite of few issues before I got the whole thing grounded properly.

[richards]

Jerry,

First, I'm assuming that you're using stock motors with an after market control box. Those motors would require 175VDC to operate at their highest potential.

If you can, change the motors to the PK296A2A-SGxx motors and wire them half-coil with a 35VDC power supply. You'll find that the motors can function much better than the motors that you probably have on your machine.

The Alpha motors are very similar to the PK296A2A-SGxx motors in power and ability; however, the Alpha motors require an Oriental Motor power supply to work properly while the PK296A2A-SGxx motors can be run from a Gecko 20x stepper driver with excellent results.

[myxpykalix]

Here is a question. How long is the power cord that runs from your circuit breaker to your bot? Or how long is the line from your circuit breaker to your spindle?

The longer the line is you will lose some power. If it is a short run, it shouldn't matter however if you have a long cord and have it bundled up under the bot, that, in combination with the added strain of the cut might be causing a decrease in the power and thus the stall. Especially if you have other tools running in the shop.
I think Ed Lang had a post about line voltage and something he did to increase it. Maybe you can get a meter and check your line voltage. If nothing else that might eliminate that as a possibility. Rereading your post where you talk about ramping up the speed and spindle and it stalling more is another indication it might be the power consumption issue.

[ed_lang]

I have not changed the power in my shop. My line voltage is what the power company sends me. I run an Alpha so my results will not compare to yours.

[jhicks]

Well, thanks guys for all the feedback. To update you and answer a few points, I am pretty sure I have isolated the basic problem as software vs hardware for the stalling. since its now happening even with air cuts, I'm in trouble and trying to get it resolved. What I've done is too lengthly to report. suffice it to say, all new wiring, new gold plated connector contacts all soldered, new water proof automotive connectors, new IGUS cable routing chain, udated windows to disable everything possible including no network connection hardwired any more even though I was only used to downlod cut files, verified point to point on every conductor continuity all along its/their path to circuit boards, restored to prior system dates, worked with Mach on their end, and several other botters directly, geckos seem ok,on and on.
Still intermitent motor stalling. ARGH&^%&%&%
working with mach software designer on every conceiveable possibility from XML files to frequency to you name it.
Called power company as PC, Inverter, vac, dust collectors are all on dedicated lines with 400 AMP 3 phase service and not much else on this main.They plead ignorance to regulating supply.
Many days now turn to weeks yet problems of intermittencies (meaning intermittent stalls) continue to exist with or without accessories on or off like vacuums, spindles, dust collectors, lights or whatever combination on multiple files from different design software one might imagine.
Usually 20 to 25 minutes into a 30 to 45 minute file. Gotta love it!!

So for now the best I can say is I wish this NEVER happens to anyone and hopefully never to me again once resolved. Next step may just be to delete hard drive and reformat for a new start but thats a last resort.
If I ever do find the root cause, I'll let you know.

[cnc_works]

I know it is a stretch, but is it possible that your cpu is heating up during the course of running the file? Just enough to cause erratic behavior? There are utilities that will allow you to monitor the bios temperature sensors and display them from within Windows.

Donn

[jhicks]

Donn, I suppose anything is possible but its been quite nice here and historically much hotter so while a good observaton, I'm about to simply crash and burn to start over again. I think they call that the "scorched earth" approach.
Thanks

[myxpykalix]

Jerry,
When was the last time you updated your control software? Maybe it is something in the upgrade? What about reverting back to an older version? Just shots in the dark.

Ed,
maybe your post was about checking the line voltages or something with so kind of meter or maybe it was someone else...I guess i'm just confused!

[cnc_works]

Jerry, external temperatures are not necessarily relevant. All computers build up dust and lint that can block the cooling of the cpu. In fact if it is bad enough the cpu can easily overheat enough to damage itself and the motherboard. I advise all my customers to check their computers every three to six months for dust and lint buildup...our shops are probably a little worse than the average house.

Again, it's a stretch, but when you are looking for an evasive problem...

Donn

[jhicks]

Thanks and agree with dust issues. unfortunately in this case dust is probably not the culprit as I open and clear dust on a very regular basis. daily if not more often. In fact my control box side panel is open so I can see and clear daily plus get improved air flow.
But thanks for the contribution and trying to help

[tkovacs]

How about an old computer repair trick: re-seat all the removable boards/connectors both in the computer and the control box. In my box the stepper controllers are in sockets so I would re-seat those too.
Did you check the power supply voltage inside the controller box? If you can, attach a voltmeter and watch the voltage while the machine is running. If the voltage changes significantly while the machine is under load the power supply may need replaceing.

[jhicks]

Thanks guys. Heres the latest. After restoring and rerunning older software and newest versions. More failures. So new pc, new software, new PC optomized. More failures.
anyone got any other ideas?
Were down to changing out the control box since thats about all we havent changed.
Any ascension boxes someone might like to sell. a working one that is?
Tomorrow a friend brings me his to plug in and we'll see.
Thanks for the support.

[Brady Watson]

Jerry,
I've had Geckos 'get tired' with limited power output, and in turn, torque. My advice would be to buy a replacement Gecko or 2 to have laying around as a spare & try swapping out where the troubled axis seems to be. If positional loss seems to be happening on BOTH X&Y, then there is a possibility that one component in your stepper power supply is shot. I doubt there is anything wrong with the Ascension breakout/mainboard. It either works or doesn't.

-B

[chiloquinruss]

Jerry - this is a stab in the dark but how about static! Could the culprit be static build up that discharges into the controller then clears itself once discharged? Just a guess? Russ

[richards]

Jerry,

If you have an oscilloscope, monitor the step pulses to the gecko stepper drivers. As long as you have step pulses, and as long as those step pulses are clean and uniform, you can rule out the computer and the breakout board as being the cause of the problem.

Next, watch the DC power supply. If you have a bad capacitor or if one leg of the bridge rectifier is blown, you're going to have extreme ripple. That will certainly affect higher speeds.

If the pulses are good and if the DC voltage is constant, then it either has to be the Gecko stepper drivers or the motors. The easiest way to check those items out is to swap in a good driver and a good motor.

Static and grounding are always a possibility, but unless you've changed something on your machine, or unless a ground cable has become loose, you'll probably be okay that way.

Personally, I'm thinking that you've got a power supply problem.

[jhicks]

Well Its quite a comfort to know that all you botters here and those who have helped via phone and direct e-mail are out there for me. I cant tell you how much I appreciate all the feedback and suggestions.
Heres the good news, I think we're back up and running. Phew!!! :-)
What was it? Well after basically replacing all the electro mechanical potentials. ie cables, connectors, power plugs and continued failures we started to ask about the computer so a new PC was set up.
We optomized and shut down non essential processes and utilities to avoid background processes interrupting mach.
Initially it exhibited the same failure mode. intermitent stalls and that dreaded GRRRR motor sound (with a Router cad file).
So we restored to original system config,(no utilities changes, just screen saver and screen set options turned off, reloaded a new updated mach version clean again,and swapped out my ascension with my friend Bob Standards ascension and developed a new test plan.
The light went on when we asked, If this is software, what could the source be? Since this PC is strictly being fed files via memory stick (network has been unplugged)we started to wonder if the actual design files could be corupt and the real problems

We design on 3 computers with 3 or 4 different software programs and each have their own XML's and post processors.
So we realized the problem had started with files coming off the enroute/router cad PC (which are the majority of our cabinet sheet cut files and volume) but hadn't really tried any older files from the other PC's or software.
The only one we had run was after things were corrupted and when we did have that failure we reloaded prior xml and finished it OK, then went back to RC files thinking they would work, but problem resurfaced.
When we tried another NON RC file from a past effective run, after a failed RC attempt and it seemed to run fine for 42 minutes and 18,000 lines of code on Thursday night on the old PC but Friday am we had a failure on a new file from my laptop? ARGH%$%$.
So we determined the new tests would be much more isolated with both of the other PC's and programs on each of them before any more RC files from the cabinet design PC. Designed identical files to run simple circle, square, and triangle cut outs and area clears to watch, listen, and learn what would happen. We isolated each PC on a separate thumb drive to insure data/pc integrity at least up to transfer/download point.
Looked like problem went away. Motors seem smother than ever in terms of noise, and vibration and no GRRRR or lost steps.
So I ran another new set of files as air cuts from a new V carve Pro file and whalla, all seems good.
So for now we dare not load any file from the router cad PC (which happens also to be one active with internet files and many other software processes)
The plan today is run a sign we are very late with and get that out of the way and maybe Monday test a router cad file (cross my fingers)
So either its something from the basic cabinet design PC embedded in the router cad/enroute system, or we've fixed it entirely. we wont know if its a complete fix, until RC files are run. But if all 3 PCs generate good files and the problem goes away, we will have to assume it was either some problem in the old PC (strange since we completely emptied the c drive and restored to original factory config) or the mach xml (also strange since this same xml was used in prior failures but with RC files).
The one thing that is still unknown is "will the problem come back in time" since we had days with 5 good sheets and then failure(s) which persisted but those were RC files from PC #3. But at the moment I'm air cutting another V carve file from PC #2 and all sounds are the right sounds so before much longer we go to the make chips mode and hope for positive results.
Long and painfull story but a HUGE thanks to everyone who either lent an ear, a shoulder to cry on, posted ideas, or shared a beer to ease the pain.
Special thanks to Brian Barker of Mach, Bob Standard of Cardinal Art design, Gary Beckwith, Gene Watson, Joe Crumley, Tom Galzin of Ex Factory, Mark Green, and Jim Smith who took my calls and helped us consider or verify many options or just were there for our "sanity checks" and other suggestions.
When we get confident enough to run files from the cabinet design PC/ router cad/enroute test, I'll drop a shorter note with the results.
For now that PC is suspicious and in quarantine as far as any cut file transfers.
THANKS 2 All

[jhicks]

well unfortunately its still a BUST. Not sure why or what to do about it but seems its all PC related and perhaps background programs interfering but whatever it is, Its not any fun at all.
Be well and dont get what i got.

[harold_weber]

Jerry, Your first note in this thread says:

"........then suddenly the Y axis stalls, loses position....."

Is it ALLWAYS the Y Axis that "stalls, loses position" ?

Also, its not really clear what control box you are running - which is it?

[jhicks]

Hi Harold, direct answer is NO its not always any axis, sometimes one, 2 or all 3 and it can be under cutting load, engaged on rack with no cutting, or with rack and pinions disengaged.
With pinions disengaged it can be close x or far x or can be one first, then the other but weve seen them all get their own meltdowns over the past 3 weeks.
Thats the nature of this inconsistent beast. Ascension control box is what we are using.
Most folks I discuss with at length say its in the PC. Virus on old, background programs getting ahead of mach on new, or some variation on that theme. When I run the mach diagnostic test I do see many pulses at constant steady level.
Had fewer pulses on the old PC and it failed as well so I'm planning to reformat the old PC hard drive and only have windows with no other programs even on it and see what happens. even though all screen saver settings, energy monitors,sound, and other control panel functions are off. Not overly confident this will be thr fix but gotta keep trying.
Seems something interferes with signals after a period of time and things go crazy but time interval isn't consistent either. Just runs ok for a while, gives me hope and crashes in a few more minutes or a few more files.

Mike, no oscilliscope but have swapped out motors early on and it doesnt isolate either a gecko or a motor, or a motor cable so that didnt lead me to the destination but appreciate the suggestion.
Power supply comes up lately also so that will be checked out soon as well.
Brady, no indication Geckos are out since failures are too inconsistent unless all 4 are intermittent. Like the Motors, whats the odds of all 4 being intermittent? Power suply is looking more suspicious all the time as well as PC perhaps.

[harold_weber]

Jerry, did you try another Ascension Box and get no improvement? If you cannot obtain another Ascension box, try hooking a DC voltmeter to a Gecko's power input terminals. Locate the meter where you can watch it while the Bot is running.

Not sure what the Ascension power supply voltage should be, but someone here or from Ascension should be able to tell you. If voltage seems OK while things are running normally, see if you can note a drop-off in voltage when you start to have problems. This is not meant to be a complete diagnostic check of the power supply, maybe Mike could help with some other diagnostics that can be done without an oscilloscope.

[beacon14]

I assume the machine is properly grounded? Many of these types of problems have been traced to grounding issues.

[richards]

You can check power supply ripple with a voltmeter. Set the meter to AC volts. The reading should NOT be more than 10% of the DC voltage (preferably 5% or less). If all of the components are working properly, there should be very little change between low stepper activity and high activity.

[jhicks]

Harold, checked input voltage at geckos and see between 10.8 and 11.0 Volts as axis is activated. next is to check direct output from power supply. I'm hearing we should see a constant supply above 65V DC so for now thats the next step in addition to complete purge of my old PC down to windows and mach with appropriate screen savers etc turned off.
Thanks again
On grounding, we never and neverhad this machine grounded but do now and problem persists.
When i added new motor cables i did use a shielded 4 condustor cable abd brought drain wire through connectors but did not terminate shield drain. problem existed long before rewire but any feedback on unterminated shield wires?
I've heard do nt ground them, ground one end, and ground/terminate both ends so not sure which advice is appropriate.

[richards]

Jerry,
Ground one end only. You should have a common ground point where all grounds are connected. If you ground both ends, you may get a ground loop, which is just another way of saying that the grounds are not equal.

On my machine, I ground off the paint on the steel frame around a hole near my control box. All ground wires and all shield wires connect to a bolt that is firmly connected to that hole.

Now, to the power supply. A Gecko G202 requires at least 24VDC minimum, so, I'm fairly certain that you've found your problem. Since I have no information about the power supply in your Ascension control box, what I'm going to write may not apply.

I'm assuming that you have a transformer, either a regular square type or a round toroidal type, a bridge rectifier, which is either four rather large diodes or a single square unit about 1" square and about 1/2" high, and at least one large capacitor.

The transformer reduces the voltage of AC electricity from your line voltage (120 or 240) to around 45 to 50 volts (assuming that your final voltage is 65VDC). The voltage coming out of the transformer is still AC. If you disconnect the output wires and read the voltage on those wires with a voltmeter, you'll probably see about 45 to 50 Volts AC.

The diodes or bridge rectifier transform the 60 cycle AC into 120 cycle half-wave AC. In other words, normal AC on a 120 line goes 170 volts positive and 170 volts negative 60 times a second. A meter reads the RMS (root mean square) voltage, which is 120 Volts. The diodes or bridge rectifier convert the negative portion of the sine wave into a positive sine wave, which is how you get 120 cycles per second instead of 60 cycles per second.

The capacitor stores those 120 per second cycles (similar to a fast drain battery). If the capacitor is large enough, it can store enough current so that the stepper motors see only a steady DC voltage level. On the other hand, if the capacitor is not working properly, it doesn't store the electricity and the stepper motors see 120 cycle AC power. They don't like that.

AC ripple is caused by the stepper motors drawing current faster than the capacitor can store it. Moderate ripple is acceptable. Some devices can tolerate a lot of ripple, while other devices can't tolerate any ripple. Stepper motors work best when there is very little ripple, but they can still work will about 10% ripple without loosing too much speed.

Although a faulty capacitor is usually the cause of excessive ripple, a faulty bridge rectifier can also cause ripple. Since the circuit is designed to have 120 fillup cycles per second, if the bridge is only sending 60 fillup cycles per second, the capacitor won't be filled up fast enough to handle the drain.

(Just in case you're wondering, you multiply the AC voltage coming out of the transformer by 1.4 to get the DC voltage. In other words, in a well designed DC power supply, the DC voltage level is the same as the Peak AC voltage level, meaning that a 50VAC transformer will produce about 70VDC.

The first time I used an oscilloscope, I thought that my brand new 'scope was broken because the 120 volts AC showed up as 170 Volts. When I called my mentor and asked him if I should take the 'scope back, he spent an hour with me and explained the fundamentals of AC power vs DC power.)

[jhicks]

Thanks again to all. So far we have confirmed its NOT the power supply, not the parallel cable, and not power cables, probably not the new PC or mach software but it is starting to look like it very well may be at component level on ascension mother board. Too early to tell but Scott Worden has been kind enough to let us borrow his control box to see if we can isolate it as control vs PC.
Transformer outout is 70V DC and consistent, switching motor cables done and confirms its not the motors but data going to them so somewhere between voltage to Geckos and motors we believe is the corrupt source and hopefully a control box will confirm or deny. Hope then we have it isolated to PC or control box. Its a nasty little headache but all you have eased a bit of the pain level with your support and helped with our sanity checks
Thanks and be happy when you're making dust.
Maybe tomorrow will tell the tale.

[richards]

Jerry,

A systematic review of all things should be the first step.

I'm a little confused about some of the postings.

Harold asked about the voltage to the Geckos at their power input terminals. You reported 10.8 to 11 Volts. Later on you said that you've confirmed that it is not the power supply. Perhaps the wrong terminals were checked, or perhaps they were checked incorrectly?

From a trouble-shooting standpoint, we always first check voltages to every sub-component. Using a common ground, we verify that every sub-component gets the proper voltage withing 5%. If a voltage is incorrect, we stop, fix the problem and then continue checking.

If the voltages are correct, we check all cables to verify that nothing has come loose.

If all cables are tightly connected, we start with an oscilloscope and verify signals. A workable 'scope costs about $500, but usually a 'friend' has one that you can borrow for a few hours. With a 'scope, you start at the signal source and watch that the signal propagates from the source to the destination without distortion. That is really easy to do with a dual-channel 'scope. Both the input and the output signal should match.

With the kind of consistent errors that you're having, it should be fairly easy to at least rule out the possible causes.

If everything checks out and the problem still exists, then it's time to verify the setup in Mach3. Is the ramp/acceleration rate correct? Have you verified that rate with the material that you're cutting, or are you using a 'standard' acceleration rate for all materials?

Solving a problem in a process controller is not difficult, but it can be time-consuming. I spent the better part of a year trying to troubleshoot my Prt-Alpha, just to find out that the problem was caused by me when I added a wireless card to the Shopbot computer.

But, in any case, using a systematic approach will help you check-off everything that is not part of the problem. Sooner or later, as you narrow the possibilities, you will find the problem.

[jhicks]

Thanks Mike, I must have mis stated something.I think Geckos are getting 60V but Sometimes with all thats being review I get mixed up. Tomorrow we have another ascension to compare against this one which will either tell us PC or ascension. at least thats the hope.

[jerry_stanek]

I just upgrade my PR to the 4g and what a difference it makes. I can now run about three times faster and my jog speeds are at 7 ips. I can take it to 8ips for jog without any chatter. The software is a lot nicer to use also. You also not stuck with the 2000 lines of code that the older DOS has. That really speeds it up as mine would stop for about 5sec. after each 2000 lines to read more.

[richards]

There are a lot of issues that the 4G solved, but there are still many issues that need to be solved before a 'perfect' solution is found.

Before anyone thinks that I'm against the 4G, please be assured that I would recommend it to anyone who is running the original stepper drivers. The Gecko stepper drivers solve many problems and I highly recommend the upgrade. If I had a PRT machine, the first thing that I would do would be to upgrade to the 4G.

With that said, I would also upgrade the motors to the Oriental Motor PK296A2A-SG3.6 or the PK296A2A-SG7.2 motors. I would wire those motors half-coil to get the fastest possible speeds with reasonable torque. Since the gear box limits the torque, I would not use the bipolar-series motor connection.

Also, I would remove the Gecko stepper drivers from the controller card and mount them on a length of aluminum at least 1/2" thick to dissipate the heat.

Finally, I would use a power supply, with those motors, that generated between 30 and 35 volts DC.

In my opinion, that would be the ideal, low-cost solution. The cost would be moderate compared to the 'Alpha' stepper motors. The speeds would be the maximum obtainable from a non-Alpha stepper motor and the resolution would be equal to or better than the resolution given by the Alpha stepper motors. (The Alpha stepper motor uses 1,000 steps per revolution while the Gecko uses 2,000 steps per revolution, meaning that a 3.6:1 PK296A2A-SG3.6 motor gives the same resolution as the 7.2:1 geared Alpha motor - although the Alpha motor has a superior gear box.)

Before the 4G came to market, some Shopbot owners switched to other brands of controllers, such as the Ascension. That controller, in my opinion, tried to address all of the problems of the pre-4G Shopbots, but because it was a generic solution, it didn't completely solve any problem. When I inquired about the Ascension 1000 controller, particularly about getting electrical diagrams and other detailed information, I was told that that information was not available to the general public. When I heard that, I stopped shopping, since I knew from past experience that all electronic controllers will need repair, sooner or later.

The point that I'm trying to make is that a controller, without the support of Shopbot, leaves the owner of the machine dependent on the tech support of the 3rd party vendor. That support may be limited or even non-existent. In other words - let the buyer beware.

If you have a Shopbot, carefully consider all the options that Shopbot provides before selecting a 3rd party supplier for any major component. If you feel that a 3rd party offers a superior product, verify that you will have access to all necessary electrical diagrams - because you will be on your own when service is required.

[jhicks]

Just a quick note on what I hope may be the final brief. This will sound strange but strange problems may be the result of strange causes.
3 things worthy of note after everything has been replaced and rewired.
1) We called the power company and a superintendent eventually arrived with his 2nd in command when we called them to determine if our power was being regulated and why we were experiencing florescent light flickering occasionally and power outages.
The supervisor walked out to the power/transformer pole feeding one the 2 mains to outside our building.
He looked up at the vines and growth into and across the transformers and said "Holy S***!
We have to get that fixed, don't have your employees in this area and don't let anyone touch that fence!
One fellow tossed a stick into the vines and generated sparks.
Upon further inspection we found a dead bird in the bramble growing up and around the area. I'll post a photo later.
2) we determined that by reducing speeds from 425 on rapid moves the GRRR diminished and eventually vanished at a slow jog of 240IPM. we felt we had discovered a symptom but not the root cause.
3) We began to focus more on poor grounding especially with the main being conductive through growth along our line and even up to the building main input conduit/pipe.
So were establishing a new ground for all the machine equipment and hoping this unusual situation it isolated and resolved soon.
Thanks again for all your support and input.

[joe]

Jerry,

I know the voltage at our shop isn't stable. But I don't think there is anything we can do about that, however there may be a voltage stabalizer big enough for 220V. That's what we need.

Hope you are on the recovery track.

[jhicks]

Met with Oriental tech engineer and I brought him 2 motors to test. He could get them to generate the "resonant frequency" grrr at very slow RPM speeds and only had a limited ability to dial them faster BUT his input was that the frequency we are sending the motors shouldnt get worse at higher speeds as the band for that particular frequency is pretty narrow. His tests never had motors stall/stop and by dialing up a very small increment, grrr went away and motors seemed to turn just fine.
So his conclusion was either its in the control
box/geckos or there is a parameter in the software
sending that frequency or whatever it is to my motors.
No way to determine exact frequency without an
oscilliscope so thats the only news I have, NOT
motors.
Shopbot may have that data but Oriental doesn't have specs locally so...

Does Shop Bot have / can I get? frequency specs on the old standard 1 amp PRT motors so we can determine suspect "resonant frequency range" and might isolate what might be causing this error when we get near an oscilliscope??

Theory was motors or software it's NOT motors maybe software, maybe geckos.

He couldn't stall the motors at very slow
speed/rpm either but mine just stop slow or fast and growl; its definately a frequency issue. Cause / source undetermined but likely software parameter or Gecko drivers in his opinion.

Not sure about geckos but 2 sets of 4 on 2 separate control systems failing at the same time just for me? HMMM doesnt seem likely but hear 202's are off the market, 201's maybe better, so who knows. The mystery continues.

[jhicks]

Looks like problems finally isolated and 99.9% solved.
Too long to really outline here but combination of software, grounding. Solution multi faceted but for those interested, either e-mail or call.
Key word is "warp9td" on the web.
Product is "smooth stepper"