Advanced Search

Good to hear that you managed to get your toolchanger working.
I would hope that the PID settings can be constant. I did think that the error term can be scaled.
We know the state of the backgear and the output (spindle encoder). With the analog output, we can have at least a good estimate of the motor RPM. So back-calculating the ratio of the variator should be possible, at least to a degree.

If we scale the error term with the overall transmission ratio and tune the PID at that state, it should give reasonably good behavior over the whole range, wouldn’t it? I am not sure of this. Doing this would give a wonky error at startup (estimating the variator would take a few rotations, I think), but it should be fine before the spindle is fully started up, I think.

But I may have an error in my thinking. Control theory is not my strong suit.

Might be worth asking in the Remora thread.

I am working on a similar solution, but want to do it as a comp, since I got my comp for the turret working this weekend.

Gear ratio can be inferred from spindle speed / commanded voltage, corrected to frequency, corrected to RPM. (You are a step ahead of me there, with your vfd frequency out put).

I expect we will need a constantly changing set of PID settings, as the inertia varies over the entire speed range.

I always wondered whether the back gear can change gear with the spindle running, but it would be pretty pointless, given the huge 6.5:1 spindle speed change, so I will only trigger it stationary.
Cheers,
Mark


 

And you see the pin iocontrol.0.emc-enable-in toggling high?

Maybe something odd with the amp-fault pin for the W-axis?
joint.N.amp-fault-in

Yeah, I would also happily sacrifice full screen for M1.

There have been recent updates to the driver. Which LinuxCNC version are you using?

@Andypugh

If you want me to subsidize the shipping cost as a contribution to the project, just let me know.  Once you get an address to @jschulze and a shipping cost is determined, I can paypal either you or @jschulze directly.

 

If you have the index connected to the 7I96S index input, then a standard
pncconf or MesaCT created hal file will have the correct connections.

jetbadger wrote:

From what I gather, logic.s0free should be an AND device, so with personality=0x105? Maybe there are other essential things there? I assume SPIN0-xx  signals are distributed in another file?

 

Yes, logic.s0free is 0x105

There are probably other essential ingredients missing from that hal file.  Best bet is for me to post my whole config...  You've got two options:

• I can give you a slightly out-of-date version right this second.  It has all the spindle/c-axis stuff in it, OR
• Wait until tonight and I'll give you the latest version
  • Includes a switchkins component for my separate parting slide, plus some changes that don't affect C-axis stuff

Keep in mind that either config is quite complicated, and there are still a number of functions/features I haven't finished:
- Collet closer position sensor logic/interlocks for both main and sub-spindles
- GUI view rotation for sub/main spindle work
- Other stuff yet to be discovered

Your pick.

The M1 button is the one i am actually missing in Gmoccapy.
The space constraint is a hard one, everybody has other needs.
For me personally i would rather skip the full screen toggle, but a total of 11 buttons would fit as well.
 

This recent commit has me suspecting that it is in some way related to non-trivial kinematics.
github.com/LinuxCNC/linuxcnc/pull/3890

I use M1 option stops all the time, to give me time to check things on a new program before continuing to the next thing. This is still implemented on my mills old Gmoccapy installation.

Did turning on/off stops on M1 really get deleted from Gmoccapy? That would be disappointing.

Today I worked on the variator.

I think I have a nice way to automate it.

The variator is set by an AC motor and has end stops at its extremes, but no way to know the current gear ratio. I have the hardware set up so the two contactors for “Go Faster” and “Go Slower” cannot be on at the same time, and they cannot be active while the corresponding end stop is triggered.

I use one of the analog outputs of the VFD to get the VFD frequency into the 7i84.

My basic idea is to use the variator to try holding the frequency of the spindle motor close to 50 Hz. The spindle PID would not be aware of this, I mean it would see the result, but the variator control and spindle PID are separate systems.

I have trouble getting the spindle PID stable even without introducing this extra error. If anyone has a rule of thumb on how to tune a spindle, I would be all ears. But I wanted to try the variator stuff anyway.

And I think it is working. I built a simple bang-bang controller, setting faster if the frequency is under a threshold and slower if it is over another. Both are reset when they cross over 50 Hz. This seems to work fine.

With how unstable the PID is at the moment, this does not really find its sweet spot, but I think the concept is sound.

I attached my current HAL and INI. This is mostly gathered from various examples. If you find any mistakes or inaccuracies that may bite me later, I would be happy about critique. The variator stuff is at the end of the custom.hal.

Thanks for reading, and happy Easter to you all.

Can someone help me understand how the cubic interpolator works? There doesn't seem to be any documentation explaining what it does. The only things I can find are for the old exact stop trajectory planner. 

As far as I can tell control.c queries the trajectory planner, runs that data through the kinematics, then creates joint->coarse_pos data and adds it to cubic.c, which is the interpolator. It seems to repeat that until the interpolator is full. 

Then it runs cubicInterpolate() which I guess takes joint->coarse_pos and interpolates it into joint->pos_cmd. 

What are the course positions and where do they come from? Are they the beginning and end of a g code move or smaller segments created by the trajectory planner?

What is the interpolator doing? It seems to be trying to create a cubic spline, but what are the way points and were do they come from? How many points does it need? 

On that note, is there something I'm missing that runs faster than the 1ms servo loop that needs position and velocity data from an interpolator?

Hopefully someone knows. The control.c program is full of comments to the effect of "this is how I think it works". 

I was not able afterall to get the C axis to behave with the previous setup. It would home and reach setpoints very well, but after going into spindle mode and back into Caxis mode it hunts for home position in almost a comical way no matter how much I played around with the PID values for the orient thingy and I couldn't wrap my head around the multiple gains for it.

I will try your config, @spumco. Ended up pretty much copy/pasting your's and atleast on the bench it starts up so I'll see if I can figure it out on the actual machine. 

From what I gather, logic.s0free should be an AND device, so with personality=0x105? Maybe there are other essential things there? I assume SPIN0-xx  signals are distributed in another file?