Print

[PhilCrompton]

I have a question for Tom or anyone who has successfully driven 12v, 0.4a per phase, 1.8 deg NEMA 17 stepper motors. I have converted an old EQ6 mount two of these with pulleys and belts.

I have been bench testing the mount for the first time with an Astro EQ that I built from Tom's PCB and pre-programmed chips.  The current limit voltage is set to 0.22v on both drivers (which are pukka Polulu).       

The problem I have is that the motors will not slew at GOTO speed if I set micro-stepping to 1/32 with micro-step gearing enabled.

What happens is that the steppers accelerate up to what would be the max slew speed but on the point of reaching max speed they make a loud screeching sound and stop moving. I am familiar with how the steppers should sound as I already have a working EQ5 with the custom pulley conversion.

If I change to 1/16 microstepping then the motors slew OK but when tracking the RA motor is very noisy. At 1/32 micro-stepping tracking is very quiet.

Am I missing something I would appreciate any pointers. Could it be that I have taken up the backlash too much and made the works shafts too difficult to turn at goto speed? The problem is apparent on both motors.

One other thing I have noticed when listening to the RA motor tracking the humming is not constant. It seems to regularly miss a beat. I don't know if this has any effect on the actual OTA tracking as so far I have only bench tested the mount.

Any help or ideas to try would greatly appreciated.

Thanks,

Phil
 

[PhilCrompton]

I have been doing some more reading on here and wonder whether my problems are related to the DRV8825 mixed mode decay.

In review, I have two symptoms:

1. If, in EQMOD, I try slewing at 800x I hear the stepper motors accelerating briefly then they stop and make a louder noise for as long as I hold the button down in EQMOD. When I release the button I hear the stepper motor decelerate.

2. When tracking at sidereal speed in 1/32 uStep mode I can here missed steps every second.

Reading other posts on here and articles elsewhere, symptom 2 sounds like I need to do the fast decay mod. Would symptom 1 also be expected from an unmodified DRV8825 running in 1/32 uStep mode?

My next problem is that the original blogpost that people link to explaining how to do the fast decay modification is no longer online. Please can someone explain what I need to do or point me to link that is still up and working.

Also, where can I find information on the "diode" modification in case I need to do this. Again, the links in the original posts no longer seem to be working.

Finally, would I have been better going for a stepper motor with a lower rated voltage? In my ignorance I just assumed that I should go for a 12v rated stepper. I have since seen a post on SGL where someone says that it is important to have a motor supply voltage that is much larger than the steppers rated voltage.

Thanks again,

Phil

[PhilCrompton]

In case anyone is following along ...

This evening I have taken the 12V 0.4A 1.8deg steppers out of the mount and tested them on the bench.

I have fixed the missed step by making the rapid decay mod to the DRV8825s by putting a solder bridge across the DECAY and FAULT pins on the ICs as suggested in another post.

The problem with high speed slews from EQMOD remains. If I try and slew at anything above 501X the motor starts to turn up to speed but then stops suddenly until the button is released at which point it seems to complete the deceleration stage as if it had been moving.

I have also done some more research on stepper voltages and I now realise that it is actually better to drive the steppers with a higher voltage that their rated value (up to 20x greater apparently according to this http://blog.inventables.com/p/stepper-motors.html).

Also, I should have realised that going for 12V steppers isn't a good idea since I drive the mount off a 12V battery so the voltage drops as the night progresses.

With this in mind I have ordered two of these (https://www.omc-stepperonline.com/hybrid-stepper-motor/nema-17-bipolar-09deg-36ncm-51ozin-09a-54v-42x42x40mm-4-wires-17hm15-0904s.html). I hope they do the trick.

Phil

[PhilCrompton]

The new stepper motors arrived. They are 0.9A per phase so I set the current limit voltage to 0.45V on the drivers and they behaved just the same as the previous motors when trying to slew in EQMOD at 800X under no load.

A friend at my astronomy society told me he had the same problem when doing his EQ5 and that he had to turn the current limit down below the recommended setting to get the motors to work. So I gave this a try and hey presto the new motors now spin up to full speed on the bench under no load. I actually ended up setting the current limiting voltage to 0.2V. I don't understand the logic to this but just in case someone has the same problem I have posted what I did to get mine working.

I will test them in the mount this evening. I'll be back if I still have problem.

[Victor_M]

Hi Phil,
one question: why do you want to achieve this "high speed" GOTO? I think 500x or 800x is too much, and maybe this is your proplem of the sudden stop of the GOTO.

I had the same problem some time ago and I fixed it just lowering down this speed (I have 400x in my mount). Doing so you get more acurrate GOTOs and you just have to wait a few seconds more to get the final position.

Kind regards
Víctor

[TCWORLD]

There shouldn't be any sudden stops anymore. I fixed the acceleration/deceleration in V7 of the firmware to be quite smooth.

[PhilCrompton]

Hi Tom,

The sudden stops after initial movement seems to go away if I DISABLE uStep gearing in the Config utility. I am using the latest version of the config utility from the links on the download page (8.05).

I managed to get the mount slewing by disabling the uStep gearing and loosening the worm backlash (i.e. at the cost of having backlash present) but I can still hear the odd missed step. Could part of my problem be that the Stepper motors simply do not have enough torque?

The ones that are in now supposedly have a holding torque of 0.36Nm (51oz.in.) the same as the SW NEMA 17 steppers in the modern EQ6 mounts (according to a post on the international astronomy forum).

My knowledge of stepper motors is limited. Is it better to have a stepper with a much lower rated voltage than 12V but higher current or vic versa?

Is holding torque important when what we are actually after is acceleration/deceleration and constant velocity without missed steps?

So far I have tried:

12v/0.40a/1.8deg/0.40Nm; and,
5.4v/0.90a/0.9deg/0.36Nm (the current ones)

Other options I could try are:

2.8V/1.68a/1.8deg/0.45Nm;
10V/0.64a/1.8deg/0.60Nm;
7.2V/1.2a/1.8deg/0.65Nm (but 6 wires and only available shipped from China)

I am loath to keep spending money on stepper motors to find out which is most suitable by trial and error. Can anyone give me any guidance to help me choose?

Thanks.

Phil

[PhilCrompton]

I have been doing some research and have come across the following page which has increased my understanding of stepper motors quite a bit.

https://www.duet3d.com/wiki/Choosing_stepper_motors

I have created a spreadsheet using the formulae given in the article for calculating the power supply voltage required for various slew rates with a range of stepper motors. See the attachment to this post.

I realise now that I have been looking at this the wrong way around. According to the article the limiting factor for a stepper motors rotational speed at a constant available current is the supply voltage. So my aim of 800X as per my EQ5 is unrealistic on the EQ6, as pointed out by Victor in another reply. The gearing would lead to the stepper motor needing a supply voltage of around 50V to overcome the back EMF due to rotation and inductance. Since my supply voltage is around 12V the best slew rate I can expect with the current motors is around 200X.

The spreadsheet does suggest, however, that speeds of around 300X should be possible with different stepper motors. I wish I had found the article earlier.

Phil 

[Victor_M]

Hi Phil,

Thanks for the update. Interesting article and good work with the spreadsheet 
Also we have to consider that at high speeds the mount will suffer more stress when accelerating or decelerating, and for sure the worms for each axis will get mechanical backslash earlier

Kind regards,
Víctor

[PhilCrompton]

Final reply from me just to confirm that I now have the EQ6 slewing nicely at 325x, no sign of missed steps or sudden stops and working with 32 uSteps and uStep gearing enabled. The figures from the spreadsheet I posted in an early reply seem to reflect what happens in practice.

If anyone is interested the stepper motors I ended up with were from OMC Stepper online part number: 17HM19-1684S. 2.8V, 1.68A (running them at 1.2A), holding torque 0.44Nm and a 0.9 degree step angle.

The moral to this story is do your sums first. Calculate the rotational speed you will need from the stepper motor to slew at sidereal rate with your gearing. Then do the maths for a range of stepper motors to find out which one gives you the best slew speed for your available voltage and current.

Anyone want my reject stepper motors, very little used but not suitable 325x EQ6 speeds?

Phil