×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

My focus values drift over time???

  • Posts: 1119
  • Thank you received: 182
What kind of autofocuser are you using? What is the setup of the stepper motor? When you say "smaller numbers", do you mean focusing inwards, i.e. against gravity?

My first guess is that your stepper is missing steps, due to current limitations and excess weight on the tube, so they need to be made up by gradually shifting the apparent focus position.

What speaks against that (and against your explanation) is that this seems to happen only with one filter. Those effects should be universal.

Are you sure only one filter is affected? If so, has it gotten loose in the filter wheel and is about to fall out? If the filter shifts position it might explain that, although I would not have thought that it would make that much of a difference.

Anyway, I think the troubleshooting has to concentrate on the fact that only one filter is affected (if that is indeed the case). Then see what is wrong with the light path affecting that particular filter only.
3 years 8 months ago #56364

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Yes, it is set so smaller numbers are against gravity, inwards. But the focuser of the telescope is super smooth. I tried finding a weight limit in the docs, but couldn't. Tried manually preventing it from moving - no way! So I doubt it's losing steps. But I indeed had not thought of that possibility.

Ah no - it's for all filters, of course, not just for one. I just compared the numbers for one filter only, as different filters have offsets.

One thing I thought of was the controller somehow forgetting the direction of the last move. But both that, and the amount of BL, is stored in the eprom of the unit, it also does the compensation when the manual control knob is used (I hardly ever do that, though).

Guess I'll disable the BL compensation for the next sessions, at least if AF can handle this. But the Linear module I use does (with my settings) a 200 steps overshoot, that should in principle be enough....
3 years 8 months ago #56365

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
Hoi Peter,

I think missed steps is the most likely reason. Having just built a Moonlite compatible autofocuser using the MyFocuserPro blueprint, I had the opportunity to find out first hand. It depends on whether your focuser is using microstepping or not, which will also increase the likelihood of steps missed, and also the current limit that has been set on your focuser. If it is operating near torque limit, the chance of occasionally missed steps will increase. That may not be apparent when you just watch the focuser in action. I am still fine-tuning mine to minimize the number of steps missed on my focuser. You want the current limit set as low as possible, but high enough so missing steps are minimized. The higher the current limit, the higher the power consumption on the stepper, the noisier it operates and the hotter it gets until it seizes up entirely. Those are the trade-offs. Went through all of them.

Check out the links below, they give a good overview of the (potential) problem and how to solve it.

Salü,

Jo

www.makerguides.com/drv8825-stepper-moto...er-arduino-tutorial/

sourceforge.net/projects/arduinofocuscon...userPro.pdf/download

cabristor.blogspot.co.nz/2015/02/drv8825-missing-steps.html
The following user(s) said Thank You: Peter Sütterlin
Last edit: 3 years 8 months ago by Jose Corazon.
3 years 8 months ago #56367

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133

Jo, and you were/are absolutely right! After your suspicion made me unsure, I looked again closely at the behavior of the unit, and noticed it sometimes didn't make noise and/or move when commanding it to - both with the manual knob and via USB. So I measured the current - and that was largely varying between .09 and .5 A. Having a closer look at the power connector I noticed a disconnect when touching it a bit stronger! :ohmy:

Next suspicion then was the connector - maybe too much play between the center pin and the plug. I had such issues before. But trying different ones didn't help.

So I finally dismounted and opened the controller box. And guess what I found? A broken solder connection of the ground phase of the DC in. I re-soldered it, now current is stable, and all commands do cause a noise/move of the motor. I strongly assume this fixes the issue, though I might better re-measure the BL, too after this.

So many thanks for getting me on the right track!

Pit
The following user(s) said Thank You: Jose Corazon
3 years 8 months ago #56368

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
:( :huh: :angry: :sick:

Now that was a short happiness. Having tested it on the desk, I mounted things back to the telescope. Neded to turn the motor, so I could access the screws of the coupler. So I used a battery to power the unit. wiggled the connector a bit, just to be sure, and something cut short. I'm quite sure it was in the plug, not in the controller. But nevertheless somethings fired now. I went back to the desk, using the PS there, but the unit wouldn't boot: Usually three yellow blinks, then, after some 5s, green. Now the blinks, then, after some 5s, again the blinks. And again....
Tried various things, when, after having been blinking for a while, the LED turned green. Hope comes up. But the motor won't move. Connected the USB cable and fired up EKOS. Everything works, I can change settings, move the position - just the motor doesn't move. Not sure how to find out what is broken, motor or the output amplifier.

Seems I have a knack for misfortune....
3 years 8 months ago #56371

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
I would measure the resistance of the stepper coils first. If you fried a coil, resistance will be high. If the coils are intact, you should find low Ohm resistance (consult the spec sheet for the stepper to find out what the coil resistance should be ). It could also be the stepper controller (not sure what focuser you are using, self-built, commercial?). Is it Arduino driven? From what you are writing, it seems the microcontroller is still functional, otherwise Ekos would not recognize the focuser. So that's good news. If it is just the motor, that is easy to replace.
Can you share the wiring diagram?
3 years 8 months ago #56372

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Thanks for the support :)
As mentioned in the first post, the unit is a Pegasus Astro DMFC (Dual Motor Focus Controler),
pegasusastro.com/wp-content/uploads/2018...dmfc_manual.v3.x.pdf
So a commercial unit, no wiring diagrams here. I opened the Motor box, but that is really only the stepper+gear in there, and a RJ45 connector.
Not sure which are the coil cables. The motor has 5, black, green yellow, red and (brown or purple). Resistances are 50 or 100 Ω, or no connection, depending on which wires I test (sorry, I'm sort of electronics analphabet...). But that rather seems to tell the motor is fine, is it?
3 years 8 months ago #56373

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
Hoi Peter,

yes, that seems to show that the motor is fine. So it looks like this is a unipolar stepper. Red should be the common wire that splits the coils in half. So when putting one connector on the red wire, ALL the others should read 50 Ohms, right? If only 3 wires then read 50 Ohms and the fourth shows high resistance, then that would indicate that the coil is fried between those two connectors.

Anyway, the common wire could have a different color, I don't know, but one of them should show that behavior. Then the other wires should group into 2 pairs that show 100 Ohms. If so, then all parts of the coils are fine and you can eliminate the motor as the culprit. That leaves the stepper controller, arghhh!

In that case, it looks like the stepper controller is fried and you would have to contact Pegasus Astro to find out what to replace it with. The PDF you provided the link to does not say.

Can you post a picture of the stepper controller that allows one to read the label? From the specifications on the Pegasus web site it looks like this may be a DRV8825. If so, that is not an expensive part to replace especially if it is not soldered onto the board but connected through a socket. Since that is the part that would have to absorb the brunt of the surge generated by the short, it would make sense that it got fried.
3 years 8 months ago #56376

Please Log in or Create an account to join the conversation.

  • Posts: 1009
  • Thank you received: 133
Hi Jo,

indeed, the black cable has 50Ω to all others, and they have 100 between them. The previous 'no contact' reports were my fault...

Yeah, I already figured I should contact Pegasus Astro, and see what they say. Attached is a photo of the board, I assume you talk about the U3 unit, close to the Motor connector? It's hardly readable, also with a magnifier. I think it reads TB713A2 6612FNG. But as you can see, no sockets :( So I strongly doubt I'd be able to replace that myself :blush:
3 years 8 months ago #56378
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
That's a shame, Peter. From your description, that short must have been inside the 12 V socket.

Well, let's see what Pegasus Astro says. The part is not very expensive. You can get one for as little as $2 from Amazon.

smile.amazon.com/Detectorcatty-TB6612FNG...s=electronics&sr=1-8

So, if Pegasus Astro says they can't repair it or it would be very expensive to do so, you may want to have a go yourself. You just need a soldering iron with a very fine tip, a magnifying glass and a steady hand. Desolder the old controller, then use fine wires to connect the main board to the breakout board.

Alternatively, I would suggest putting your own autofocus controller together. This one works beautifully.

This browser does not support PDFs. Please download the PDF to view it: Download PDF

3 years 8 months ago #56380
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 1119
  • Thank you received: 182
Hoi Peter,

I sent you a PM, let's continue discussion offline.

Jo
3 years 8 months ago #56390

Please Log in or Create an account to join the conversation.

  • Posts: 2
  • Thank you received: 0
Sorry to bring up a 1 month old thread...

I am curious what the final outcome was here? Did Pegasus fix the unit for you?
3 years 7 months ago #57815

Please Log in or Create an account to join the conversation.

Time to create page: 0.949 seconds