Print

[Washy]

I just posted this at the EQMod group also as I didn't know how long my account would take to approve, but it was fast

I've been pulling my hair out all day now, and have given up for the night now

I've got an EQ5, I've mounted 1.8deg/step motors with 14T pulley, the worm has 52T giving 3.71 : 1 so I've punched that into the AstroEQ configurator, my AstroEQ is a Mega2650 clone and 2 DRV8825s.  The worm is 144:1 because in my desperation I've even counted that.

The issue is in EQMod if I go from the park position 90dec around to 0dec indicated by the control panel the mount only moves 65 real degrees, or going from 90 to -90 the mount moves 130 degrees.

I've got the feeling the problem might be my wiring but I can't see how a problem with the stepping mode wiring could give anything other than 2x errors, where mine is ~38%.

Also when sidereal tracking the RA motor doesn't seem to pulse smoothly like maybe a third of a second micro stepping followed by two thirds stationary maybe or maybe not related.

Not sure where to go next with the debug I thought AstroEQ told EQMod the right drive train parameters so no setup was required to get the speeds correct?

My motors are NEMA17, 2ohm, 24v rated although I'm running them 12v and 0.6vref - they shouldn't stall or skip steps in this application...

Any pointers gratefully received ☺️

Clear skies!

[TCWORLD]

Hmm, that's very odd. You are correct that AstroEQ should (and does) report it's gear ratios to allow EQMOD to work out the correct speeds. It's odd that it is moving only 38% of the required distance.

Your gear ratios seem correct. So it seems there are three options:

• The motor is skipping steps - it's possible, especially as the motor is running well below its rated voltage, but seems unlikely given how consistent it sounds
  
• The mount parameters are incorrect. Maybe you entered the wrong motor step angle?
• EQMOD isn't correctly reading the parameters - I've seen this happen before if there are somehow strange custom mount settings in the EQMOD ini file. It may be worth deleting that file (%APPDATA%\EQMOD\EQMOD.ini) and letting EQMOD recreate it. Then try running again. This will however cause all of your other settings in EQMOD to be lost, so it might be worth backing up the old file first - if clearing the settings doesn't fix the problem, you can restore your backup file.

Hopefully it's one of those options otherwise I have no idea what could be happening!

[Washy]

Thanks for your rapid reply - I don't think its skipping steps either... There is a video of it slewing here...

https://www.youtube.com/watch?v=CxVqLv8Mhxk&feature=youtu.be

I tried re-intialising EQMOD like you suggested but that didn't help either.

I was wondering what frequency clock does your firmware expect from a mega2650?  I didn't know if clones had different clocks sometimes?  I'll re-open my box tomorrow with the electronics in a double check all the wiring and the mega.

Here are the AstroEQ configurator screencaps and EQmod also...

[Washy]

Of course in the morning its blindingly obvious why its not working.

So I write a quick arduino sketch to full step the motor round 360 degrees and it works AOK, its 106971.4 full steps but 32x that != the numbers in AstroEQ they are 39% different

my calc = 3423086   
astroeq calc = 2457601
% diff = 1.392856576

Now I can't explain that - but I hit the update button in AstroEQ and the calcs on both axes became good and the scope now moves as intended   Happy days!

I guess I didn't realise I had to click the update button on the configurator 

[TCWORLD]

Ah well, glad it's solved  .

It should update automatically when you click done, but that bit of the utility is a bit glitchy - it's something I've fixed a bit in the new version.

[Washy]

Yeah the 38% difference is of course the pulley config you used and put in the custom pulley config file and mine.

I tested everything out last night and primarily it all works nicely!

 I still get the same RA behaviour when tracking sidereal the motor only moves for about 0.3 - 0.4 seconds per second - it tracks the sky well but I guess it might be noticeable if I used a long focal length??  Is this the decay mode issue?  Or is that just the normal run cycle?  Maybe I'll put a scope on it later and measure more precisely the on/off timings....

[TCWORLD]

Sounds like it's the decay mode issue on the DRV8825 showing up. There is a thread around here somewhere on that.

[Washy]

Yeah I've just read through that - I put my drivers into fast decay and they run smoothly and continuously now - just a bit pitchy.

Hopefully that will improve my imaging as running a pulse a second or so made my stars go a bit soft.

[Washy]

This was a god send, I've just transitioned to a RaspberryPi setup and it wasn't working right - wanted to try upgrading the firmware which worked but bricked me eeprom - so I got my ratios back from here since the computer I originally used is long dead and no back up and I didn't fancy counting teeth today much!!

I input all the same data, but it wouldn't accept the same 808 rate as before so the params computed all different - I uploaded it anyway and am just doing a tracking test to make sure its now working properly.

Sidereal tracking still wouldn't engage when pressing start in ekos but after manually slewing the mount and trying again it appears to work just fine now!