I just got a Burny DNC 2.8 and...

More
04 Feb 2022 08:54 #233956 by tovarna
Hi all. I am Javier, from Spain. I just bought an old, 1988 machine from Cleveland Machine Controls.
It says BURNY and DNC 2.8
Still have not plugged it in, but I am sure it would pay off if I just change all the electronics (if I can keep the servo motors).
So I just started by collecting all the info, and first thing that comes to my mind that a Mesa 7176E and a thcad might be the way to go.

It has an Hyperterm 100A...but I have not even checked it up. 

Any opinion on this idea? it is the way you would go?
I know CNCs (well...one I have), but it is the first time I get hands on a plasma CNC. Not afraid of learning curve and programming, but I know little about electronics. Maionly I am wondering if I can use the current CMC dc servomotor MT2630-126AF, enconders H20DB-37-SS-60-ABZC-28V-V-SC18, and the drivers with a Mesa717E, but I am openminded and would like your comments and expertise.
Thanks 

Please Log in or Create an account to join the conversation.

More
04 Feb 2022 12:40 #233966 by tommylight
DC servo motors, so most probably use +-10V analog control, not step/dir.
So a Mesa 7i77 with one of 7i92(eth) or 5i25(PCI) or 6i25(PCI-E).
Also there is Mesa 7i97.
First try to start and jog the machine to make sure the drives/motors work properly.

Please Log in or Create an account to join the conversation.

More
11 Feb 2022 12:17 - 11 Feb 2022 12:35 #234565 by tovarna
Replied by tovarna on topic A week later...
First of all, thank you Tommy for the answer.

I am waiting for the electrician to properly review the machine and set a plug for it, but in the meantime I went for a mesa 7i95, since it can use servos and stepper, and I planned to use the old ones if they work and then implement new Z axis with stepper.
I got it to work to work with a raspberry pi4 with linuxcnc 2.9 over ubuntu (as a test), just following the threads about the 7i95 and I just made a stepper to move and now testing limit switches.

So I will keep on learning how .hal files work, and how to set up inputs and outputs (probable a spindle for drilling holes in the future), but the number one on the list is to control the plasma and for that I have seen the THCAD10 board, but also a THCAD 5 for ohmic sensoring? Would two THCAD10 work? They are same price here, so...just asking. I have a Hypertherm max 100 21

Will go back to forum reading but I wanted to thank your effort and time spent already answering lots of questions and giving advices.
Last edit: 11 Feb 2022 12:35 by tovarna.
The following user(s) said Thank You: tommylight

Please Log in or Create an account to join the conversation.

More
18 Feb 2022 18:25 #235258 by tovarna
Replied by tovarna on topic A week later...
We just got one motor to work, and we will try to change the brushes to the other.
Trying to understand the decoder and the hal code. 
Any ideas on the thcad options I asked?
Thanks

Please Log in or Create an account to join the conversation.

More
18 Feb 2022 20:32 #235271 by tommylight
Replied by tommylight on topic A week later...
2 of Mesa THCAD10 should work just fine, but it would be better to get one and make the machine work with a floating switch first, then you can better evaluate if you need ohmic sensing or not.

Please Log in or Create an account to join the conversation.

More
18 Feb 2022 20:43 - 18 Feb 2022 20:43 #235274 by snowgoer540
Replied by snowgoer540 on topic A week later...

Any ideas on the thcad options I asked?
Thanks


Save yourself the hassle and skip THCAD ohmic sensing.

Phill did a nice write up on THCAD vs Relay ohmic sensing (and you can find a schematic to use the relay method there, which you can use pncconf to set up very easily): here

The setup is way less complicated, there is way less fiddling to get it to work, no components to track down in this forum and install, it's cheaper, and it's not bothered by noise/interference, and those of us who have ditched THCAD ohmic have found relay ohmic to be way more reliable, etc. It just works, right out of the box.
Last edit: 18 Feb 2022 20:43 by snowgoer540.

Please Log in or Create an account to join the conversation.

More
18 Feb 2022 21:19 - 18 Feb 2022 21:20 #235281 by PCW
Replied by PCW on topic A week later...
I suspect the reason THCAD ohmic sensing was unreliable for you is that your
setup has higher leakage currents than others. A relay circuit would have a
_much_ higher current threshold than the THCAD circuit people are using here.

If you have higher leakage currents, a THCAD based ohmic sensing circuit
would need a lower value sense resistor (with appropriate power ratings)
Last edit: 18 Feb 2022 21:20 by PCW.

Please Log in or Create an account to join the conversation.

More
18 Feb 2022 22:26 #235294 by tommylight
Replied by tommylight on topic A week later...
Also forgot to mention, ohmic sensing requires a shield on the torch, it can not be used with torches that have only the nozzle on the outside.
Unfortunately i never got the chance to test ohmic sensing as i can easily do a relay one and a THCAD one (for once in my life i have 2 of THCAD10 and a THCAD300 at once :) ), but due to all my torches having no shield, i can not. That nozzle is used to fire the torch so it can go up to multi thousand volts on every torch start.

Please Log in or Create an account to join the conversation.

More
19 Feb 2022 00:45 #235299 by snowgoer540
Replied by snowgoer540 on topic A week later...

I suspect the reason THCAD ohmic sensing was unreliable for you is that your
setup has higher leakage currents than others. A relay circuit would have a
_much_ higher current threshold than the THCAD circuit people are using here.

If you have higher leakage currents, a THCAD based ohmic sensing circuit
would need a lower value sense resistor (with appropriate power ratings)


That could be true about the leakage current, although from the THCAD-Ohmic standpoint, I ran the same setup/components as everyone else here, save for a Hypertherm power source, and my setup wasn't the only one to exhibit the same issues.

The relay circuit ran fine with the same resistor recommendation from the THCAD-Ohmic circuit. We moved to a lower resistance values only to make it be more water tolerant. This is something THCAD-Ohmic requires as well, it can't be fixed via a moving average algorithm in the component.

All that aside, in the end, after a TON of testing with both circuits, and now months of experience with both circuits, we found that there is no benefit to THCAD-Ohmic (yes, I know it is "less parts"). Sure, it "works", and it's an alternative, etc. But I can't see a good reason to spend the extra money, to add the extra complication (I am not talking part count), and frustration necessary to make it work. A quick search will reveal the many who struggle to make it work.

I suspect that the reason it's made it this far is due to the ultra-sexy marketing buzz word name it was given... "hypersensing".

Please Log in or Create an account to join the conversation.

More
19 Feb 2022 01:51 #235302 by rodw
Yes hyppersensing was  a beautiful marketing buzzword I invented. Too bad there is nothing to sell with open source :).
Strange, its worked for me  for years now. it also gives much quicker response than relays which by definition are slow.

I got caught today with some kind of short in the tip but the ohmic test shows that up and its not restricted to hypersensing. In fact it was added as a requirement from John  Moore who contributed the relay circuit.

One thing I've noticed is that QTplasmac does not handle such an issue gracefully. I'm awake to it now. It blindly tries to lift the torch to break the circuit but if its at top of travel it can't as it tries to go out of bounds. maybe it could issue a warning if a job starts in that condition or raises an error if the ohmic sense is on and the Z axis is above probe height at the beginning of the probing sequence. Would need to think through the best approach.

Please Log in or Create an account to join the conversation.

Moderators: snowgoer540
Time to create page: 0.185 seconds
Powered by Kunena Forum