Stock Electronics Settings and Calibration Process

Well, I now have sled, frame, motors, power, software, it’s all looking good, and making some funny jiggling sounds when I attempt to move motors, so I’m guessing I’m close now.

I’ve scoured through lots of info pertaining to issues folks have had with calib and setup, but haven’t managed to find definitive calibration and settings info. Is anyone able to point me in the direction of such a resource if it exists. Ideally, I’m after a step by step for calib, and the Ground Control settings for the stock motors.

It’d save me lots of time and I’d be very grateful - pouring virtual beers as we speak ;¬)
Cheers,
Miles

1 Like

Congrats! It’s a relief to get the machine to that point. I remember such a tremendous feeling of accomplishment myself.

Maybe I’m a little confused by the question but there is no “standard” settings, even for the stock design. There are enough variables in construction that the auto calibration process is really necessary to get accurate measurements. There are a couple of settings, such as motor steps per revolution, teeth count, chain size, etc that all should be correct right out of the box. Just make sure that you are using the latest release of the firmware and Ground Control (v1.02). Simply follow the instructions that guide you through the auto calibration process, and let us know if you run into any problems.

On a side note, it may be prudent to double-check the measurements the Maslow is making with a tape measure. For instance, when the machine is doing the distance between motors test, have GC tension the chain, then run a tape across the chain to make sure that the distance that the Maslow sees is in fact the same as the actual measured distance. In the past, this method has helped catch problems before cuts start to get ruined.

One of them beers would be nice, though. :stuck_out_tongue: Best of luck with the calibration. Hope to see projects coming off your machine soon!

2 Likes

Thank you for getting back to me so quickly MM, I’ve just donned my Bar stewards apron and replenished your virtual pint ;¬) It definitely is a milestone getting the build complete, but I’m holding off on the celebratory beverage until I’ve at least got the The Mas drawing circles with a Sharpie! Might be a while before I make it State side to get you a real beer, but if you’re ever in the UK maybe that could occur sooner.

I started out with some Alibaba motors, so have messed about with the GC settings and now don’t know what was standard - will look for a reset/defaults button.

The instructions you mention that guide the auto calib process - do they appear when I hit that button? Sorry, I’m conscious of sounding like a button fearing retard! Do I calib first without the sled attached, or hook everything up and go from there? I’ve found forum threads alluding to a Step process for calib, I’m guessing this is the auto pop up thing. Heading to the garage to investigate with a trigger happy mouse finger!

With thanks
Miles

2 Likes

I second this check of the motor-to-motor measurement, and would add that making the same measurement when there is no tension will tell about whether the frame is flexing under tension :confused:, a source of inaccuracies.
Edit: Mine’s a soda…

1 Like

Much thanks! Beers and work make for an interesting combo. Just don’t tell my boss! xD

Definitely on my short list for places I want to visit. We’ll see how quickly the trip is realized though :wink:

That explains a lot. An easy way to reset all your settings is to delete the groundcontrol.ini file in your user directory. (C:\Users\ [UserAccountName] for Windows) The next time Ground Control starts, it will make a new ini file and use the default values. I forget what the actual default values are or I would post them here so you could double-check them. @blurfl, @krkeegan, @bar, would you happen to remember what the settings should be with stock hardware?

I know it’s daunting, but let me assure you it’s pretty easy. And don’t worry, there are no dumb questions. Hitting the “Calibrate Machine Dimensions” button begins the process, and instructions for completing the procedure will show on your screen.

The first couple of steps will need to be done without the sled attached. I believe the instructions will tell you when it’s time to attach the sled to the chains.

That is a very good point! I totally forgot to mention that. Thanks for adding that :smiley:

2 Likes

1.03 which comes out on Wednesday will add the default values next to each setting to make remembering what the defaults are easy! One of the many excellent changes. The values are auto-populated so they will stay correct even if the default settings are changed in the future! Here’s what it will look like:

The order I would test things in would be to click “Actions → Test Motors/Encoders” to make sure your motors are working right. Since they are different from the stock ones a wire could be crossed somewhere. If it all looks good there “Actions → Calibrate Machine Dimensions” should walk you through the calibration process

3 Likes

Thank you all for this help. Have to confess when I hit the test motors and they both performed, a protracted yelp of excitment left me ;¬) Despite I’d connected them L-R backwards. That remedied, I initiated the calib dims guide. I am now running on a pair of motors that Hannah arranged and had sent to me - many thanks Hannah, they arrived safely and swiftly.
I’m finding that the CW and CCW buttons don’t seem to work consistently. I’m guessing this is because of an encoder issue? When testing motors they perform smoothly, but a 1 deg CW or CCW seems to make them jiggle, and not always move on. I have encoder steps set to 11.0, am I good to go on that? And if it isn’t that, is it likely PID values? (something I know nothing about).
Thank you,
Miles

2 Likes

That does sound like an encoder steps issue.

Are you talking about this encoder steps setting is 11?

That number is the encoder steps per revolution of the output shaft so it should be the PPR of the encodergear box ratio4. The four is because we’re using quadrature encoding and we actually get 4x the resolution.

It might be worth taking apart your gearbox and counting the teeth because we found that the ratios on some of the gearboxes out there are not actually what they are advertised as

1 Like

That is probably caused by PID settings being too aggressive, try reducing the K
value

the motor testing just runs them flat out to make sure they spin, it’s not
trying to move them a specific distance.

1 Like

Right, as I’m working with the correct motors I’ve changed the encoder value to that described in Bar’s post above, it seemed a good starting point and easier than demounting motors and opening the gearbox. This seems to have remedied the jiggling the motors were doing before, YAY!
However, now I’m getting the incremental 1 deg movements, followed by a higher pitched whining noise that carries on for a few seconds following the movement. It sounds worrying, should I be concerned, or am I fretting unnecessarily?
Might try and reduce the PID K value to 4 and see if that helps.
Cheers,
Miles

1 Like

I’m completely fresh to the whole PID thing, and don’t know what I’m doing. I’ve found the PID K settings, but don’t know which to play around with, or what I’m trying to effect. Is there a previous thread describing this element of the system?
Cheers,
Miles

not really, the K factor is how aggressive the system is in trying to move to a
new location, and the other factors affect how it reacts when it overshoots.

the stuttering (and probably the whining) of your motor is probably caused by it
trying too hard to be precise and overshooting (and therefor having to back up)

I believe the K value is a small integer (6 or so), try reducing it a bit. it
will make the machine accelerate a bit slower, but should stabilize things.

values in the 3-5 range have been mentioned.

you want it as high as it can be without makeing the machine unstable.

That is totally normal. The motors will hold position for two seconds after each command and you are hearing the sound of the motor holding position

The noise you’re hearing I’ve heard described as the encoders “settling” on the final position. I hear the same noise every time my machine moves and nothing has caught fire yet :stuck_out_tongue:

I don’t have a great deal of experience with PID tuning, but I have been able to tune my quadcopter and rover autopilots in KSP successfully. Each of the values determine how the machine adapts its movements to get them as close as possible to the desired location. Simply put, the more off any of these settings are, the more erratically the machine will try to compensate for being out of position.

If you’re running the default settings, these settings will work for you, although the encoders may be a little noisy. After you start cutting, you won’t be able to hear the encoder noise over the router and dust collector.

Thank you for the appeasing info MM.

Once the Maslow motors became available I jumped on them and ordered 4 (I’m using one for my z-axis, and wanted a spare). So,didn’t have to demount from the frame, and have as such cracked open the gearbox of the spare and found the following gear ratios:
Twin start worm (Brass) --> 25T Nylon Compund 10T --> 35T Metal Compound 9T --> 29T Metal Compound 18T --> 37T Metal Output gear.
2:25
10:35
9:29
18:27
This corrolates with the thread on gearboxes Whats inside the gearbox/motor?
Have to say, I think the quality of the Maslow supplied gearbox is much better than those I sourced from Alibaba.
So, Guess I have the right encoder res at 8113.73, and the gearbox is good, so maybe I’m just a bit paranoid about the whining noise ;¬)

3 Likes

It’s getting late and it’s close to zero degrees in the garage so I’m going to call it a night. But I have now initiated the zeroing of the gears and shall complete the calibration tmrw morning.

Thank you all for your help today. I’m totally blown away not only by the Maslow concept, but also this wonderful community. Never will I feel lonely again in my garage! :¬)

All the best,
Miles

4 Likes