Advanced Search

I was faffing around with Linux a bit and managed to get a wifi driver working (hurrah) and then I thought to myself, why isn't PathPilot recognizing the 5i25T if it's supposed to appear to the OS as a "5i25". You were right PCW that it has a different PCI ID and path pilot will not normally recognize the card. 

PathPilot's card decision logic is in a python file shoved in a bunch of folders. It figures out if there's a 5i25 installed by matching the PCI ID for the 5i25 with the output of lspci. If that python file doesn't find PCI ID (I think this is right, recalling it by memory) 2362:5125 it falls back to checking the ethernet connection. Simply changing the logic to look for 2362:5925 (PCI ID of the 5i25T) instead allows Pathpilot to move forwards to booting. 

The next problem I run into is that I believe the HOSTMOT2 PCI driver is supposed to flash the card installed with the firmware provided by Tormach. I didn't find mesaflash installed on the system so I assume the driver handles the flashing process itself. I know that mesaflash requires you to specify the T when flashing the card or it just returns "card not found" and hopefully it's the same with the driver. Going to dig through more files to see if I can find where it's being referenced

It seems like if you modify the setup python file for the card you could pretty much have it recognize any Mesa card as the logic isn't that complex. I could probably write up a little guide on how I changed the python files to recognize a different Mesa card than supported if you think that would be helpful? I don't see a lot of information on the guts of PathPilot in the forum.
 

You rock!

I noted your comments on Z axis machining and thought I might stick my neck out here.
Your machine looks like it came with the old Philips controller from one of the early photos. They set the axes up differently to what I believe is the current convention. That is with the Y and Z axes swapped. So you might find your servo closing relays in the wrong physical order on the DIN C rail. The convention seems to be X side to side, Y forward and aft, and Z up and down. Certainly Heidenhain uses that on controllers from my TNC 155 up to 400 series. Early Bridgeports used the TNC 155 so they are that convention also.
Rotary axis A, B and C rotate around X, Y, and Z axes respectively. So I code my rotary axis as C, but it is possible to not use that convention by changing a parameter.
 So in the your programming examples you wouldn't change the axes orientation, you simply tell the machine the cutter in in the Y axis by using G18 in the G code.
There is a program showing this in one of the links to my cloud previously. It's the one where it pretends to cut a spiral thread using a cutter in the Y axis and the C rotary table.Rather like a vertical lathe.
The next problem is that on our machines the cutter is usually attached to the Y axis, so it moves that way. The other 2 involve the table moving up and down and sideways.. On the typical Bridgeport of that era, the quill moves the cutter in the Z direction and the table moves back and forward and sideways. And yet you set up the program identically for both!
So the convention is that the DATUM is the work piece. So when you do the familiar right hand rule with your real hand, You place the back of your hand on the work piece whilst standing in front of the machine. And keep it stuck in the same place on the work piece as it moves.
Now look at where the cutter is moving relative to you hand:

If it is moving in the direction  your thumb is pointing then it's moving positive X
If it is moving in the direction your index finger is pointing then it's moving positive Y
If it is moving in the direction your middle finger is pointing then it's moving positive Z

That works for all types of 3 axis machines.I have no idea how they label machines with 5 axes and seemingly infinite rotation axes.

In our case what seems counter intuitive is that for X+ the table will move left and for Z+ the table will move down.

Cheers,
David
 

Where is the ISO and does it boot in Live mode?
All my Ryzen's are gone except a R7 5700U in a laptop with a TDP of 15W including GPU, so should be worst case scenario for testing.

Won't irqaffinity=0 help?

Can i have 20 of those, please?

Here are the pages from my manual on how to disassemble and re-grease the spindle. Mine is NT40 so a bit bigger and with a different pull stud. But I would think much the same. I've added the page on how to adjust the claws. It's very quick and simple. The whole assembly item 2 is fitted when it's all assembled into the spindle, in my case back in the machine. It simply screws in. When the depth dimension is right ( I used the tail end on my dial calipers) you simply tighten the grub item 1. That presses the ball into the coned recess in the stud and expands the thread onto item 2. The hatched portion where item 2 points to are actually longitudinal cuts in the thread.
Your 30 size tool holder will have a different adjustment length I would think. .In the diagram you can see that as the claws are retracted they are jammed closed by the bore of the spindle. That means that it is more likely any rattle is due to damaged or worn claws I think. The claws are the finer cross hatch.
Cheers,
David

Tuning?
wiki.linuxcnc.org/cgi-bin/wiki.pl?Tuning_LinuxCNC/HAL_PID_Loops

forum.linuxcnc.org/10-advanced-configura...to-example-mesa-7i77

gnipsel.com/linuxcnc/index.html

Manufacturer name and model numbers for the servo drives, servo motors and encoders, also upload your current ini and hal files. 

Thanks for the info guys. They are no longer building the 7i77D as you have all sussed out - so Ill use the 7i77 with a bit of extra logic. I need the X6 servo interface the 7i77 gives and Ive got the current 7i77D built in so Ill pursue that path for the mo . Ill check out the 'U' options though for the future
Thanks again..

Working with a 7i97 board
Cmc drives and dc brushed servo motors
original drives have tac but the encoders are wired directly to the 7i97 board. 

problem I’m having is I’ve only been able to jog forwards. The 7i97 board is sending either plus or minus 10v like it should be but for some reason I only get one direction. 

I’ve switched the signal wires around to see if it would make a difference and the motor still only jogs forward. 


I’ve switched back to the stock 7i97 hal and ini file. Only changes I’ve made to it are 
ip address 
Commented out the z axis since I don’t have one yet. 
frequency I’ve changed to 5000 to match what the manual said. 


Been working on it for a month or so with the same results. Digging through the original machine book only thing I can find that catches my eye is there’s something about uniswitching (not sure what that is) when the machine is using high speed. And h bridge when at low speed for torque. 

I don’t know if that matters or not. I’m lost. 

There are at least three ways to generate a bipolar analog voltage from PWM


1. Using PWM/DIR (typically using an op-amp circuit that
is changed from inverting to non-inverting by a analog switch)
(this has the disadvantage of a glitch when direction changes)

2. Using the PWM generators UP/DOWN mode and summing
a positive and negative filtered voltage.
(this has the disadvantage that the filter time constant
varies with the input)

3. Using the PWM generators offset mode (so 0 = 50%) and an op-amp
filter that has a DC offset of - full scale
(This has the disadvantage that you lose 1 bit of resolution, which may not matter
if you use dithering)

Found one with bad I/O (pin 13)  on P1 but Tormach only uses P2:

store.mesanet.com/index.php?route=produc...t_id=411&search=7i92
 

tommylight wrote:

Does it have an RJ45?
Supposing it does not as none of mine have it, so get an USB to RJ45/LAN adapter, around 7 to 15$.
I have 6 of those adapters from very old 10MBit to new 1GBit ones and all work perfectly fine with Mesa.
And i have a Dock for the Y370 with all the bells and whistles, got it for 30 Euro used!

 

From some quick Google research, I think your Y370 is likely Gen2 (2017), while mine is Gen1 (2016), hopefully that won't be a problem, conflicting information about USB ports on the Gen 1, not sure if it has the fast USB-C port, I think not appears to be , 2x USB3.0 and one USB2.0.
USB3.0 is pretty fast, so will probably be okay if I get a decent Ethernet dongle.

Would like to see your USB to Ethernet dongle, I don't want that to be a bottleneck for the MESA card.

Thanks! Yeah, things are going well, just wrapping it up. Latency is a bit high for me but I ran `rtla timerlat top` and some of the spikes are coming from my Radeon GPU. Others are coming from APIC and hrtimer (CPU interrupts, can't tune these out.)

I'm on a Ryzen system and I've noticed that PREEMPT_RT has much lower latency on some Intel hardware. It'll be interesting to see how latency performs on other hardware with the image.

I just pushed a huge update to my portage configs on GitHub, almost ready to publish the updated installer and root filesystem:

github.com/NTULINUX/gentoo_backup/commit...9333804131fbc352c3d5

@spumco did you ever get this sorted and working?