Showroom My 3 DOF seat mover + GS-4 + Simvibe

Yes. I understand what you are saying but I don't think that is the problem because I am not going outside of the feedback range.

That is wrong. It is there to give you a little extra travel past 100% of travel. Not of their target. If you are not going over 100% then it is not a problem.

If you outside of the feedback range you will get a "Feedback Disconnect" error.

• Bit 5: Feedback disconnect
  This error occurs when the feedback is above the Absolute maximum or below the Absolute minimum (these parameters can be set in the configuration utility). The absolute maximum and absolute minimum can be set using the configuration utility.

This only happen if you go past 100% of an axis movement. Changing the Absolute min/max only gives you a little bit of leeway if you overshoot past 100% of travel.

When these motor driver error occur I am nowhere near my Min/max so changing the Absolute Min/max won't fix those errors. If I go off track I am maybe use 30%+/- from center when the motor driver error occurs.

I very really go to 100% of the motors movement so I am not exceeding the feedback range.

If you look at the screenshot I posted earlier I had 2 "Feedback Disconnect" twice over a whole night of testing. So yes. changing those absolute Min/max will fix those 2 errors but not the 2224. The 2 feedback exceeded happened on one really sharp corner, you can see it on my video I posted.



The feedback just sets the range of movement which is fine. It can't cause there motor driver errors. Even if I did go outside my feedback range they have functionality that returns it back, You can even set a different duty cycle for if you do go out of range so you can get it back in to range quicker.

You are showing your utility screens most of the time with a connection lost error, and hence it "blacks out" what may be missing.
Sometimes feedback errors can be disguised at serial protocol errors or can contribute to the issue.
After 4 years of working with these things I hope I have a bit of an idea.
Rob and Nick have suggested things as well, especially Rob as he is the JRK king.
Forget the error occurrence count. If it works , ignore it. Even when my motors are steady, the occurrence count is working overtime.

The connection lost is just because I didn't have my simulator plugged in at the time I took those screenshots. all the settings stay the same when you unplug it, they just get grayed out so you can't change them. Those are the same settings I have when they are plugged in. It is just easier for me to move my rig out of the way while screenshotting and posting.

My rig is on furniture sliders so I move it out of the way when I am using my desk for work or writing these long posts. It is easier for me to write these sitting at my desk using my nice mechanical keyboard rather than sitting in my rig using my wireless keyboard.

My GS-4 isn't the most comfortable seat to sit in so I prefer sitting in my office chair if I can.

So please ignore the connection lost error. It just means that I have moved my simulator out of the way so I can type this. All the Motor driver error happened while I was doing driving earlier in the night.

I wasn't meaning to question your knowledge, which I greatly respect. I was just pointing out that I was not going near 100% travel so adding some extra to the absolute feedback range wouldn't help my errors.

Since then I have also found something else out which is skewing the error count. When I hit my kill switch, it causes 20-50 motor driver errors. This would probably be a dying gasp type thing as the JRK's continue to drive the motors after the power has been cut.

This means that the actual amount of errors that I am getting while driving would be much lower.

At the end of the day, while on track my motion simulator is working perfectly. The PID looks great, if you see my video it is responding very nicely and has enough power to keep up with the movements.

ok, sorry if I came across wrong. As long as it works, then when you can , fiddle with PID and motor settings.
Maybe try adding boundary to the forces to "level" out the big jolted in forces.
Goodness knows I have spent a year on my personal ride still trying to get it right.
Dave.

As long as the JRK is suitable heatsinked/cooled it will fault less at the '0' current setting. As mentioned by everyone else, you although really need to find out what is causing those 70A spikes. I would go back and check everything, especially the balance of the seat. Do you have a video of the motion that causes it to fail? A 200W motor with a 50:1 gearbox should drive that seat if it is balanced fine. I also notice that you have a seat tensioner. Don't forget to bias the balance of the seat to take that extra load into consideration, especially with 85mm arms.

The JRK's are very well cooled. They have huge heatsinks and 2 fans blowing at full speed. It it is really loud but I went totally overkill on the cooling.



I will do some videos of it tonight.

@Avenga76 great and professional work! You are using a relay ?

Thanks. A friend of mine built the controller box for me.

The relay is for the kill switch up by my wheel. I have 2 kill switches and I didn't want to have the full current going up to the front of my rig so the front kill switch triggers the relay and the other kill switch on top of the controller box which is wired directly in to circuit.

This means I can always reach a kill switch.

@SeatTime The seat is well balanced. This problem even happens when the seat is not connected to the lever arms.

Here is a video of the errors. This is the easiest and most dramatic way of showing the errors. If I was to drive off road then they are very hard to spot.

This is moving the motors rapidly about 50% of the total travel using the JRK utility. I did this so you can see that it is not going out side the feedback range.

I also disconnected the seat from the lever arms so you can see that it isn't a seat balance problem

Listen for a rapid clicking noise near the end and you will see the motors slowing down. That sound is the JRK's switching the motors on and off because of this motor driver error. There were 79 motor driver errors in this video. A have a shot of the errors page before and after and the only errors I get are the motor driver errors.

This isn't a very good example of real world use because the motors are never moving that much while racing. but it does show the errors well. As I said before, on track is fine it is just when the motors need to reverse a lot like off track.

Wow, I think you may have a stuffed gear box or motor, it should not be doing that at all with no load! That noise is certainly the JRKs shutting down. Do both motors do this?

@SeatTime yes, both motors are doing the same thing.

For comparison, below is the type of weight (approx 18Kg) that I was able to reliably drive with a 200w motor with a 60:1 gear box. So you have something very wrong going on with those motors/gearbox.

Nice, Yea, I know overkill..

Yeah, that Intel cpu heat sink is totally overkill

The motors are fine if I give it a little pause between each movement.

And it took about 25 quick movements and then it failed briefly then a worse string of errors only after a few more movements then an even worse string on the next movement.

It is like the JRK's can't handle it for very long, almost like they are overheating but they shouldn't with the heat sinks and fans I have. It is like the problem gets worse and worse the harder and longer I push it. While I am on track it is okay because I don't need to push the hard enough to make them fail, but if there is a series of chicanes in quick suscession then it can be enough to cause a small string of errors. Off track pushes it too far and you get errors pretty quickly.

During this testing I almost thought I had lost a JRK. It wouldn't let me move the motor. At all, every time I tried to move to motor I got a pop up window saying it couldn't set the target because of a Comms error. The it came up with another pop up saying firmware error.

I had to unplug the USB and kill the power to get the JRK's going again.

I am a bit worried I might kill the JRK it I try to replicate that Comms error message again. That was using larger movements which made the problem much worse.

You've definately stumbled upon a Very Unique problem Rich. Im sure we can get it solved.

Whats the P and D again?
Whats your current motor power settings?
From the video you get errors, but it still moves. At what point does the JRK actually shut down?

Since you people are showing cooling overkills, this was mine back when I was building my sim. I was scared of JRK's overheating with big motors so I've cranked it up a notch

Fan inside the fan - Fanception

I really hope we can get it solved. It has be a bit stumped at the moment.

The JRK's didn't completely shutdown in this video. The normal symptom is that clicking noise and the motors slowing down as the JRK's error out. During the testing last night I did have a complete crash of the JRK's but I didn't capture the message on screen. I don't know if I want to try recreating the total failure of the JRK's again in case I blow up my JRK's for real.

When the errors happen in the video that motors slow to a crawl as the JRK's are switching the motors on an off, In racing this causes one of the motors to stay in the same place until the JRK's have recovered.

Here are my PID and motor settings

Maybe they do just get that hot that quickly.

The JRKs should not be getting hot when the motor is not under load, my guess without being there is that the gearbox is binding and creating its own load when it tries to change direction quickly. This is the only thing that would make sense.

@SeatTime
How about this for a test.

If I remove the motor from the gearbox and just try just spinning the motors with nothing attached to them. No gearbox or anything.

If it still fails then we know it is something in the motor itself or the JRK which isn't working.