Rotary table indexer

Quote:

---

Originally Posted by dnaman

I expect I'll end up reprogramming the Pic, or buying another, but I have a job set up on the mill and wondered if perhaps Steve could comment based on his knowledge of the firmware. I did look for a way to reset everything in the hopes that everything would come back but I can't find one.

---

I haven't checked but I think the version of the firmware you're using has "Set Worm Ratio" rather than "Steps for 360".
If so then it assumes a 400 step motor such that a 90:1 ratio would give you 90*400 = 36000 - which is the value you'd enter in the later version of the firmware.

I changed it because setting the number of steps is more flexible and it's obvious if you exceed the limits (entering 180 as a ratio gives the false impression that it's valid when in fact it internally overflows since 180*400 = 72000 which is greater than the max value of 65535).

Quote:

---

Originally Posted by kwackers

I haven't checked but I think the version of the firmware you're using has "Set Worm Ratio" rather than "Steps for 360".
If so then it assumes a 400 step motor such that a 90:1 ratio would give you 90*400 = 36000 - which is the value you'd enter in the later version of the firmware.

---

Yes, it does have the "Set Worm Ratio" and I didn't notice that that was *instead* of the "Steps for 360". That being so, and since I already have the worm ratio set for 90, I'd better check the dipswitch (or jumper - can't remember) settings.

I notice that you changed the 18F452 to an 18F4520 at some point. Are these interchangeable on the early PCB (when we were making our own) or do they need the latest PCB?

Quote:

---

Originally Posted by dnaman

Yes, it does have the "Set Worm Ratio" and I didn't notice that that was *instead* of the "Steps for 360". That being so, and since I already have the worm ratio set for 90, I'd better check the dipswitch (or jumper - can't remember) settings.

I notice that you changed the 18F452 to an 18F4520 at some point. Are these interchangeable on the early PCB (when we were making our own) or do they need the latest PCB?

---

The 4520 is a direct replacement for the 452 so you should just be able to swap them.

The firmware has been fairly stable for a number of years now, most changes have been fixing edge case bugs or minor changes to deal with the way people use the controller so generally if it works I wouldn't bother changing it.
(My main indexer still has V1.2 in it! I do keep meaning to swap it but there's never been a need.)

Thanks, Kwackers. While I was mulling it over, I decided to go ahead and buy the new, programmed 4520 and, since there were some changes to the PCB I went the whole hog and bought the full PCB kit. It'll get me up to date and I need a little electronics project for the Winter (I have mechanical projects coming out of my ears). And the price is right!

Quote:

---

Originally Posted by NNEast

I had emailed him a copy of the Driver Data Sheet. This is what he saw: If all of the Dip Switches (For the Microstepping) are set to ON then that is the default for the motor - 360 degrees divided by 1.8 = 200 steps ......

.......Nowhere on the data sheet does it say: "Leave switches to ON for native mode of the stepper motor" or anything even remotely alluding to that state.

Thanks again.

---

Can you expand on that NNEast? Those two statements would seem (to me) to be contradictory. i.e. there was nothing on the data sheet to even remotely indicate that all switches should be left on - yet your friend discovered that fact when you sent him the data sheet.

Please believe me, I'm not being picky - I just seem to have a problem in this same area and I believe I follow the data sheet for my driver but would like to understand this.

Quote:

---

Originally Posted by dnaman

Can you expand on that NNEast? Those two statements would seem (to me) to be contradictory. i.e. there was nothing on the data sheet to even remotely indicate that all switches should be left on - yet your friend discovered that fact when you sent him the data sheet.

---

I must admit I can't see anything on the data sheet that says what happens if you leave all switches 'on'. My best guess would it would default to 'full' stepping which would mean 200 steps per rev of the motor, but that's an assumption.

In this case it's fairly straightforward though, simply set the dip switches for 400 steps.

Do you have a data sheet for yours? What problem are you seeing?

Quote:

---

Originally Posted by kwackers

Do you have a data sheet for yours? What problem are you seeing?

---

I do, but it's also marked on the driver housing itself. I got back to this after along absence and found the switches set for 20 microsteps (motor is 200 steps). I had also set the worm ratio to 900 (actual=90) I must have thought, at the time, that the two (mis)settings would compensate for each other - I realise now that they wouldn't.

When I set the microsteps to 2 and the worm ratio to 90, the motor runs poorly and will stall. I now realise that my driver isn't up to it (max current=1A). Don't know how I came to use that. I have a higher rated one on order. (In fact I have a higher rated, and probably suitable one, already but it's rather large and won't fit the room I have in the custom case I built - which I rather like and want to keep).

Quote:

---

Originally Posted by dnaman

When I set the microsteps to 2 and the worm ratio to 90, the motor runs poorly and will stall.

---

1 amp in itself shouldn't be an issue - and a bit odd that the motor runs when microstepping. More likely I'd have thought that the voltage to the driver is low?
(FWIW I run my 2.5A motor at about an amp on a 19v supply)

I use a 5 volt 1 amp motor at 19 volts and 1 amp setting on the driver with a 36:1 worm ration with no problems.
John.

Quote:

---

Originally Posted by kwackers

1 amp in itself shouldn't be an issue - and a bit odd that the motor runs when microstepping. More likely I'd have thought that the voltage to the driver is low?
(FWIW I run my 2.5A motor at about an amp on a 19v supply)

---

I thought that at first (I was using a 19V 2.6A supply) so I switched to a 24V 3A supply. It didn't really help. Even with a loose engagement of the worm the motor still stalls at one point in its rotation .... presumably the tightest spot in the worm drive. When I rotate it by hand, I can feel this "tight" spot but only barely. It's easy to turn and you couldn't call it abnormally stiff by any stretch of the imagination.

It's a NEMA23 1.8deg 125 oz.in 2A per phase bog standard motor which should easily drive it. I will check the voltage at the driver but since the supply is wired direct from the input connector to the driver and I've tried two supplies, I don't expect it to show anything.

That leaves the motor itself. Limiting the current to 1A won't help (I^2 x R ?). The motor has "been around" in various devices in my shop so it's possible it's the worse for wear. I do have another similar motor I could swap in to try it but it has a single-ended shaft and I really want the double end to attach a manual handle. These double ended versions seem to be becoming hard to get.

Or perhaps the driver has a problem.

This may take a little while - I have to go in for some surgery in a couple of days (nothing major - pacemaker replacement ....battery run out) but I won't be able to be very active with my arms for a couple of weeks. I wanted to get this job on the mill done before that so I'll probably put the RT back to manual this afternoon to that end.

Quote:

---

Originally Posted by gandalf1

I use a 5 volt 1 amp motor at 19 volts and 1 amp setting on the driver with a 36:1 worm ration with no problems.
John.

---

I don't doubt it at all. I'm sure I ought to have plenty of oomph with my setup. What size and holding torque is your motor? Also, what RT size are you using (mine is 6")?

The table is home made using a 36:1 worm and wheel. The wheel is about 2 3/4 inch diameter and the motor is 60oz in. The actual table top depends upon what I want to do but at present has a 5 inch 4 jaw chuck fitted but since I only make small parts those sort of diameters would only be when drilling holes on a pitch circle. A thought you must have considered yourself then. Have you tried running with the motor disconnected from the table? If not then by gently squeezing the end of the motor spindle you should feel and variation in output torque but from the East side of the Atlantic it sounds like either a bent worm shaft , bent motor spindle or possibly both.
John.

I haven't disconnected it but your suggestion did give me an idea. I loosened all the screws assembling the motor mount/coupling and gradually tightened them evenly while checking the rotation. That cleared the tight(er) spot and it is now even easier to turn by hand with the motor disconnected. So easy, in fact, that I can't believe that any self-respecting motor of that size wouldn't eat it up and spit it out.

In fact it does turn the table now well enough for me to complete the job I was doing. I only need to turn the table between cuts (not while cutting) and it managed that except for some missed steps at one stage which let me know that it's still marginal.

I'll swap out the driver when it arrives and try it with this and a replacement motor.

OK, I tried something else. With the indexer driving the motor, I tried to stop the motor by hand (I have a small handle attached to the shaft extension). Not particularly difficult. I then connected another motor (same spec) and tried to stop that - I couldn't.

Seems the motor that's installed isn't giving anything like its rated torque. Which would explain my problem.

Another separate (I think) issue. As I said I have a handle attached to the motor shaft extension to help in manually setting up the rotary table. I've noticed that, if I have the motor connected to the indexer but not it's power-supply, when I turn the motor by hand the LCD panel backlight lights up. Presumably the motor is acting as a generator. Is this likely to do any damage? The motor is attached via an easily released connector so it's easy to avoid but I'd like to know anyway.

The stepper should be already protected, because the stepper creates back EMF, from the magnetic field collapsing every step it takes.

Quote:

---

Originally Posted by dnaman

I've noticed that, if I have the motor connected to the indexer but not it's power-supply, when I turn the motor by hand the LCD panel backlight lights up. Presumably the motor is acting as a generator. Is this likely to do any damage? The motor is attached via an easily released connector so it's easy to avoid but I'd like to know anyway.

---

Absolutely yes!

Most motors when manually rotated will generate a voltage that can exceed safe limits of the circuit and backfeed, destroying unprotected devices. With most motors normally a snubber diode would be used to re-circulate the current so it stops backfeed, but this isn't practical in a lot of cnc applications as this would cause current to continue flowing.

Connect your multimeter to the stepper leads and spin it, you might be surprised by the level of voltage generated.

cheers, Ian

I've been using this excellent design from one of the supplied kits for a couple of years now with no problems. But just got a question stimulated by discussion over on the ME forum. When dividing by a number "N" which isn't an integer divisor of the number of steps per rev, how does the controller deal with the resulting error? Does it use something like the Bresenham algorithm to minimise the error.

Sorry if this has been asked and answered before!

Thanks, John.

Quote:

---

Originally Posted by JohnHaine

I've been using this excellent design from one of the supplied kits for a couple of years now with no problems. But just got a question stimulated by discussion over on the ME forum. When dividing by a number "N" which isn't an integer divisor of the number of steps per rev, how does the controller deal with the resulting error? Does it use something like the Bresenham algorithm to minimise the error.

Sorry if this has been asked and answered before!

Thanks, John.

---

Hi John,
that's exactly how it works.

So basically internally it multiplies the number of divisions by the actual division its on then divides the result by the number of steps for a full circle - this gives it the nearest number of steps it needs to get to that division (doing the multiply first prevents loss of resolution although it means it needs 32 bit maths which is why the max steps per rev is 16 bit).
Internally it knows exactly how many steps it's currently taken so simply adds (or subtracts) that from the answer above to get the number of steps to take.

This means there isn't a constant number of steps since it may need to take an extra step (or lose one to be accurate).

In practice the accuracy of the maths is half the single step resolution, so a 400 step driver on a 90:1 worm gives you 0.01 degrees per step so the accuracy is 0.005 degrees. (Mechanical accuracy not withstanding).

Hope this answers your question.

Cheers,
Steve.

Brilliant, many thanks Steve.

I know this has been asked before but I couldn't find a good answer what would be a good stepper motor and driver for a 10" rotary table