Advanced Search

Those words were said cos I couldn't understand the point you were trying to make.

I don't get what issue you're having with ise.

Are you sourcing /opt/Xilinx/14.7/ISE_DS/settings64.sh before doing anything.

Now what I'm going to say is coming from someone who has worked 95% of his life as either a builders' labourer or civil labourer (roadworks). I dropped out of school at 15 and the closest I've come to any trade qualifications is 3/4 of a carpentry apprenticeship.

I've installed ise 14.7 webpack on a fresh install of mint 21.3 a week or 2 ago on a laptop belonging to rthorntn.

Installed webpack and installed a licence on my bookworm Linuxcnc box this morning, and have been using it for an easy 12 months on my main machine, running Mint 21. This particular machine has been building the mesa firmware I run on my boards.
To say it's trash is utter BS, just cos you can't get it to work is not representative of the software. If you want help to get it setup convey your errors and I'll try to lend a hand, but it sounds like you just wanna put shit on it cos you can't get it working. That's a YOU issue not a ISE issue. I don't think you're a bad person, I just think you're wrong about the software.

openFPGALooader & Sprtan6 boards.
openFPGALoader works fine, tho you really need to read the docs to load the firmware into the SPI chip. It's a 2 stage process that loads a temp bitstream into sram then that config is used to write to the SPI chip (ollie's makefile doesnt address this and it wasn't what I'd call an issue ATM, I just modded the makefile). Ollie's makefile only loads into sram and you lose the config when you remove power.This needs to be done externally from the IDE.
I add this to the MakefileI wouldn't copy and paste that as make does like it's tabs in the right place. This is the bit,xc6slx9tqg144, that tells openfpgaloader to which temp bitstream to load.



For whatever reason openfpgaloader can be a bit flakey on reading the SPI id's and can be a PITA, I think the issue I was having in reading the SPI ID on a certain chip was that it may not have been either 100% error free or it was not genuine. Which kinda surprised me as I bought it from o Aussie guy off ebay. Another issue could be flakey du pont connections, yes we all know better.



I will admit getting the Xilinx DLC9LP USB programmer to work is a pain as is the parallel port programmer (which is a genuine Xilinx product).
The tools aren't rubbish, and just cos they are a few years old doesn't make them stop working.

I've also set rthorntn up with a Spartan6 dev board and a couple of daughter boards, but due to his choice of breakout board I had to design a buffer board as all the IO is via optos on his chosen breakout board.

In all honesty I think old mate is just spending a lot of money to he doesn't have to pay what it would cost to ship a Mesa board to Oz.
This is the second project I've given him assistance with.
Whilst I haven't tried the Gowin hardware route I've messed around with rio on a couple of ICE40 boards and several different Spartan6 boards.

In all honesty I think it would be a waste to give up on your design. I'm gunna guess you've put a good amount of hours routing the board. It would be silly to throw the towel in.

Try:

I came across subroutine example to set my Z probe. But first I wanted to simplify it in order to understand the subroutines.

I have added the following to the machine ini file

[RS274NGC]
SUBROUTINE_PATH = macros

[MACROS]
MACRO = touch_plate_z

and have made a file in the macros directory that have the following statement


o<touch_plate_z> sub
G54 G0 x0 y0
o<touch_plate_z>endsub
M2

what statement I need to enter in the MDI window to execute this subroutine

Good day

Philip

 

Thank you all

Yes I would say that I am happy with the performance of the motors. 

The reason I am asking is that I am building a second machine, and in the process of deciding on the power supply.

My best regards

Philip

Good day

I have built it directly from source with no custom modifications. The last git commit is as follows:

commit ec2dc5fe7d1b55bb43c5f96b1e2f31e3d61d2fa8 (HEAD -> master, origin/master, origin/HEAD)
Merge: 8eea44d784 84f5e89ad1
Author: Greg <26096779+snowgoer540@users.noreply.github.com>
Date: Sat Mar 25 06:54:38 2023 -0400

Working link from above
forum.linuxcnc.org/30-cnc-machines/29358...-boards-mesa?start=0

I'd stick to the spec of the drivers. Unless you want to live on the wild side. And if you are using an unregulated PSU I'd be even a little more careful.

The manufacturers' put them there for a reason. They don't pick random numbers to make life difficult.

To say that heat is usually a killer isn't the full story, why is the heat being generated ?
Over voltage ? Which can lead to over current.
Over current ? Which can lead back to over voltage.
Short ? Which can lead back to over current.

Sometimes components, for whatever reason, just fail.

7I97T (plus maybe other PWM servo configurations?)
is missing the connection from the PID output to the PWM value.

That is for example:

hm2_[MESA](BOARD).0.pwmgen.00.value

is unconnected

Obvious missing bit (in red)

net joint.0.output <= pid.y.output
net joint.0.output => hm2_[MESA](BOARD).0.pwmgen.00.value

Don't forget!

github.com/NTULINUX/RTAI/blob/master/README.BIOS

This link might be helpfulhttps://forum.linuxcnc.org/30-cnc-machines/29358-proto-trak-mx3-retrofit-which-boards-mesa?start=0
 

If you are not getting the speed and power with 48vdc supply you can use 60vdc. Just make sure the drive heat sinks has good contact with your panel/frame. What usually kills electronics is heat. However, If you are happy with servo motor performance, stick with 48vdc supply.

I originally couldn’t connect at all to the 7i97t till I updated mesaflash to 3.5.9 and installed mesact 2.1.4.
I also tried changing the hm2_eth static address but I don’t think that worked. 
​​​​​​​
 

My drivers for some reason rated for 58 volts DC. 2 volts short of the standard power supply which is 60

I am using 48 volt power supply, thinking that any higher voltage may damage the drivers.

Can I use 60 volts DC power supply ?

Thanks

Can you post the hal/ini files made by mesact?
That would help diagnose the issue so mesact can be improved

With a pncconf created file, make sure you set the PWM offset mode true:

setp   [HMOT](CARD0).pwmgen.00.offset-mode 1 # offset mode so 50% = 0

 (for all PWMgens, not just 00)

Also for best ripple/resolution, follow the  PWM guidelines in the 7I97T manual.