Kflop Digital IO for Relays

[a_j_p3]

I have a kflop+kstep system that has been working pretty good the last few months, but recently the digital IO stopped triggering some of the relays I have. I checked the code and tried swapping out relays, but the issue persisted.

Eventually I found the culprit: the relays operate on a 3V signal, but what I am finding is that sometimes the digital IO output is only 2.9V or 2.85V, and when this is the case the relay won't operate... So it's pretty hit & miss recently if things function.

Are there some kflop IO pins that are more or less prone to this? I.e. could I expect better results if using different pins? Or do I need to find more suitable relays that can handle lower signal voltages? Or is this low of a digital IO voltage abnormal and possibly I have another issue (I am measuring voltage after about 6" of wire after the kflop connector so I don't think it's cable length)?

[a_j_p3]

I thought I had it sorted out, got the relay to work again by reducing the outputs from the kflop 3.3v source... But now I continually get 'Read Error, no Motion Devices connected' everytime the relay triggers and the plasma torch turns on...

Everything was working great, then the voltage dropped to the point I could only power the one relay, but now I have to restart the device everytime I attempt to turn the torch on... It is no longer a useable machine.

I checked to see if somehow the torch lead (what some people refer to as torch ground) somehow has made a connection to the control box Gnd, but it has not. All wires are shielded and everything else is inside a steel enclosure (plus it's a blowback start, not the HF start that generally cause the need for shielding, etc). I'm not sure if it's a Kflop issue or if something I did slowly created a kflop issue, but at the moment I'm inclined to believe it is one of those 2 things...

I have another kflop running another machine, I could theoretically swap it into this enclosure and see what happens... But I'm worried I may end up facing the same problem in a few months unless it's truly just a random kflop error that all of a sudden started happening.

[a_j_p3]

Ok so relay issues is resolved with a separate 3.3V supply. I have not had to do this for past kflop projects but seemed to be the logical solution and does work.

As for the read failure and disconnect issue. That is still not solved, but I have eliminated the Kflop because I swapped it out for a different board and cannot tell any differences between the two. I did check for ground loops and add a USB isolator, which helps. But now I'm inclined to think it's actually a computer issue because a few of my read error/disconnects today were accompanied by a windows blue screen error and system restart. it only happens when the plasma torch fires up so I still am assuming this is a noise issue? I'm just really confused why it just began happening.

[TomKerekes]

Regarding 3V signals: TTL outputs are only guaranteed to go to 2.4V under load, but they usually go higher especially with light loading. The first 8 IO on JP4 and JP6 have 150 Ohm pull down resistors so will not go as high as others. You might wire the relay to 3.3V supply and have the output go low to turn it on. The relay should then always see more than 3V. With the before mentioned IO with pull down resistors there may be an issue with the relay going on before KFLOP initializes them. This may or may not be an issue depending on your use and if you have some other global disable.

How was the separate 3.3V supply connected and used? 3.3V should not be fed into KFLOP.

Regarding disconnects: Sorry to hear this. If it occurs when the plasma torch "fires up" it sounds like noise or grounding. You might determine if KFLOP is resetting/rebooting or if there is a USB error. One method of determining this is to turn off a KFLOP LED. if the LED comes back on then KFLOP re-booted.

Grounding is complex. Without knowing many details about how things are arranged and wired its hard to help.

[a_j_p3]

Thanks for the response Tom. Kind of a junk show over here but figured it's easier to post than not, maybe someone else will encounter similar and with any luck my situation will improve.

Regarding the separate 3.3v supply, I will try to explain. Initially I had all of the limit switches pull 3.3V from JP4 and JP6 - the connection was: 3.3V from JP4/JP6 into a NC switch and back to a JP4/JP6 I/O (with a capacitor filter on each I/O input). Additionally the relays used JP4 and JP6 3.3V as their source power, and received their signal trigger from a JP4/JP6 I/O. The relays are 3V but came with a breakout board, cheap, apparently draw < 20mA, and I am running 5 of them in total. Now what I am doing is using the separate 3.3V supply to provide the voltage source for the limit switches and relays - now the only thing JP4 and JP6 provide power for are the relay trigger. Also for reference, the kflop power comes from separate 12v and 5v supplies, not the USB.

This separate 3.3V supply is not directly connected to the kflop, however it is providing the input signal through the limit switches into the JP4/JP6 I/O. I assumed this was OK, let me know if that is incorrect.

I will investigate your suggestion with the kflop LED to see if it is rebooting, but my gut tells me it is not since it maintains it's position without needing to re-reference the axes. The parameter requiring re-referencing is stored in a persist variable which I've only ever seen be cleared when re-booted in the past. But I will confirm.

I hope it's something like a USB issue, grounding can be real complex for me to digest, especially on this system where the torch clamp is in contact with the steel frame the machine is built around... I don't know what that may introduce but it's unavoidable mechanically to operate otherwise with this system (and likely all plasma systems I've ever seen). Totally understand that's difficult/impossible for someone else to help without physically being here. Still appreciate the insight.

Will update more when I get around to diving back into this (got what I needed to cut out today done and need to move onto other things at the present time).