Unit values or something else?

[chipster]

As I continue to update / upgrade / fix / set up my new (2001) PRT standard, I've run into some issues, that I'm hoping for help with. FYI, having just bought this, I have zero Shopbot experience.

I just connected my new 4g controller and put on new 20T pinions. As I moved my bot around, trying to become familiar with it, I notice that when (as an example) I tell it MX1, the outcome is more like MX3. I'm guessing that the concern is the unit values following the upgrade. Or, could it be something else?

I looked up the values in the user guide, and tried those figures without accuracy. Would anyone have any suggestions?

XY motors are PK299-01AA, 20T pinions, 1.8 step
Z motor is PK296A 1A-SG7.2, 20T pinion, 0.25 step

I'm using version 3.5.18, and as mentioned above, 4g controler.

If I'm reading things right, the manual says X and Y should be 458.3664 (times 4) and the z should be 1833.4656

[ken_rychlik]

Here is the link that I used when putting the 4g board on my 2004 prt. There are unit values, based on 25 tooth pinions though. That may not apply for you.
http://www.shopbottools.com/files/SBG00306061116PRT4gUpgradeInstall.pdf
I hope it helps.
Kenneth

[chipster]

If the 458.3666 value for X and Y for the 20T pinion is multiplied by 4 (per the manual), you get the same value as is listed in the link for the 25T. I tried that number, without success.

I don't see the 2291.831 Z value anywhere in the manual.

[Brady Watson]

Chip,
Your XY motors sound like they are direct drive (1:1), and therefore, the UV for them should be 3.6 times less than a standard PRT with 3.6:1 geared motors, plus 25/20 (5/4) the UV for the difference in 20T vs 25T. So...

1833.456 (values for standard 3.6:1 PRT w/25T pinion) / 3.6 = 509.2933

509.2933 X 25/20 = 636.6166 UV for XY

Your Z axis has a 7.2:1 motor, which uses the same 20T pinion as a 3.6:1 Z motor. Your UV will be exactly 2X the 2291.831 for a 3.6:1 motor or 4583.662 for the Z.

One thing to note - the 1:1 motors don't have as much torque or resolution as a geared motor, so some types of cutting will suffer.

-B

[chipster]

great help Brady. I'll get out there and try that.

I'm curious about the direct drive aspect. How do I determine that? I was just researching on the Oriental web site trying to understand these motors. The picture for that model number shows direct drive, where as mine are offset.

I very much appreciate your and everyone else's help. It's as important to me to understand the "why" as it is to understand the solution to my problems. I figure that way, when this comes up for someone else in the future, maybe I can be the one to help them

[Brady Watson]

Chip,
If your motor is in fact a PK299-01AA, then it is a standard stepper motor with 200 discrete full steps per 360° revolution. PRTs used Allegro 1/4 stepping drivers, which essentially gave a 1:1 motor 800 1/4 steps. The 4G uses Gecko Microstepping drivers, which step at 1/10 steps per pulse, or 2000 steps per revolution. A gearhead stepper, such as your 7.2:1 SG (straight geared - has a decent amount of backlash) would be able to resolve movement down to 7.2 X 2000 or 14400 microsteps per revolution.

Just for comparison, the smallest angle that the PK299-01AA could resolve is 360°/2000 = .18° (hence the 1.8° per full step or .18° per microstep) and the max the 7.2:1 could resolve is 360°/14400 = 0.0027° per microstep.

In terms of mechanical resolution of the PK299-01AA, the best it will do is (assuming a ballpark 1" diameter for 20T pinion):

C = pi X D --> C = pi X 1" --> C = 3.1416"

3.1416" = 360°
X" = .18°

X/3.1416 = .18/360 ---> X = .00157" resolution per microstep with 1:1 stepper

The Z axis with 7.2:1 would be .00157"/7.2 = .0000218" per microstep.

0.002" per microstep (rounded) is the best mechanical resolution that you will get on the X & Y axes, which, as I said before, may not be enough resolution for you if you plan to do cutting in dense materials, cutting very small parts with a lot of detail or other intricate operations, including v-carving.

If all of this doesn't make sense now, it will later


Understand that you are not necessarily 'stuck' with those 1:1 motors. If you are feeling industrious and resourceful, you could follow in the steps of Mike Richards and make yourself a similar type of belt reduction to gain some resolution on those X&Y axes. You can read up about it here (http://www.talkshopbot.com/forum/show.cgi?tpc=312&post=39324#POST39324), bearing in mind that your XY motors have a 1/2" shaft, not a metric one like Mike's Alpha.

-B

[chipster]

Thank you Brady, It worked! According to my tape measure, it's dead on. Once I get the vac and the dust collection hooked up (darned spiral duct suppliers!)table built, I'll double check everything again with actual cut pieces.

I was mistaken about my motors. The XY are direct, and the 7.2 Z is the indirect motor that I had thought about.

This sure is an adventure! I just skimmed your earlier post. I sure appreciate all you did. I'll dig into it more this evening. Now, it's time to leave to visit my 93 year old Grandmother! I'll bet my excitment about my SB will be missed by her!

Thanks again!

[Brady Watson]

Anytime Chip. Part of owning a ShopBot is knowing that others are willing to help out when you get stuck.

-B

[richards]

I agree with everything that Brady said. A 1:1 motor, driven by a Gecko stepper driver through a 1-inch pitch diameter spur/pinion gear will give you 0.00157 inches per step. As a point of comparison, I just measured the thickness of a sheet of yellow legal pad notebook paper. A page is 0.002 inches thick.

So, in a perfect world, your cuts should have cutter marks spaced 0.00157-inches apart (assuming you've selected optimum chip-load settings). I doubt that you see those evenly spaced tool marks. The imperfections in a piece of mechanical equipment are often much larger than the electrical/mechanical resolution of the stepper motor/stepper driver.

The greatest advantage that I gained when I added the 3:1 belt-driven gearboxes to my PRT-Alpha seemed to be related to the increase in torque at standstill. One of my pet theories is that the holding torque on a 1:1 motor, when the stepper driver enters reduced current mode, is insufficient. Multiplying that stopped-state holding torque seemed to be a key element in getting much better cuts.

Later, when I took advantage of the upgrade pricing and bought three 7.2:1 geared motors to replace the belt-drives, the cut quality remained visually identical to the quality that I had been getting with the 3:1 gearing. After close examination with a magnifying glass, it appeared that much of the imperfections in the cut were identical to the imperfection in the rails. In other words, when a V-roller ran over a bump or ridge on the rail, that bump or ridge on the rail caused an identical bump or ridge in the cut. There were unexplained variations, particularly at points in the cut when a second axis started moving. Those types of errors were attributed to the flex in the machine and to the laws of physics that describe what happens when a large mass starts to accelerate from a stopped state.

Anyway, I long ago stopped trying to get the perfect cut. Now I concentrate on keeping my machine properly greased, properly cleaned and properly maintained so that the post-cut sanding will be minimal. Sanding seems to always be required. When my machine is properly maintained, I might have to use 2 to 5 strokes with the sandpaper. When I've let things go too long, it might take 20 (or more strokes) to even out the edges.