As this Cutter is still a work in progress with a lot of testing, experimenting and of course learning, i am using tap water for cooling (bad idea, i know).
I’ve already had high voltage spikes in the water bucket as the Pump is not grounded, so as the rest of the wohle water tubing isn’t.
These HV spikes (>1000V) traveled through the Pump (12V DC) to the psu (simple Plug PSU w/o PE) to the SSRs. (This is the other reason why they gone crazy)
As i had a Temp Sensor clamped to a metal Tube in the Coolant Tubing, i think i somehow grounded the Water but not reliably.
Now i just connected an alligator clip to the Grounding Point at the Cutter and everything works well. It also worked well if the Pump was disconnect from Power.
I’m still not sure where the HV in the Water comes from. The Tube is ok. I think about ionisation or static charge.
Because my digital multimeter goes crazy (LCD Flickers and Beeper made strange sounds) when measuring Voltage but when measuring Current absolutely nothing is noticeable (because grounded to PE through the internal shunt.
@Marian_Scheja the HV is present in the coolant because tap water is to CONDUCTIVE … maybe :). It would be interesting to know if you still need this ground using distilled water.
This is an interesting finding. Most of us use submersible AC aquarium pumps which are not connected to a DC supply anywhere. I have wondered about using external DC pumps to electrically isolate the entire water path. In your case the DC pump did not isolate the coolant?
In your case the HV present in the coolant is somehow electrically connected to pumps DC power and conducts back though the power brick and into the control electronics by way of the AC power. That is surprising? I wonder if there actually is an arc occurring inside the pump?
The lesson here to insure that the coolant has a solid path to ground that does not include any power or electronics.
Should we be grounding our buckets to the case???
Where is this voltage being measured: “Because my digital multimeter goes crazy (LCD Flickers and Beeper made strange sounds) when measuring Voltage but when measuring Current absolutely nothing is noticeable (because grounded to PE through the internal shunt.”
@donkjr Got it working! It took moving the Arduino Uno a couple feet further from the LPS and tube, and it suddenly behaves great! I’m going to do some more testing to if I can isolate it down directly, but for the moment I’m content. Thanks for all of the help!!!
@Alexander_Flick glad you made progress but that is surprising and suggests its radiated? Did you change anything about the wiring in the process of moving it?
@donkjr Nope, nothing else was changed. I’ve been trying to further track down the source of the issues, but still no luck. Even having it set on my laptop away from everything isn’t a 100% “fix”, as it still occasionally resets itself. Last night I couldn’t get the dang thing to fail (even with it next the the LPS/laser/etc!), but then tonight it was having issues even sitting away from them. Bizarre.
I did notice that I too have some HV in my water bucket (AC pump, distilled water, but it’s been sitting there a bit now, so it’s probably picked up some stuff). I had preemptively grounded it, but wasn’t really sure it was doing much. Took the connector off the ground wire that runs to the water bucket tonight and shorted between that connector and the wire that runs to the bucket with my finger on accident and felt a little shock similar to what static electricity shocks feel like.
I’ve got a new HV power supply that I hope to try out tomorrow - that should at least confirm that the current one is fine or that it’s the source of the problem. Starting to think that maybe I should have just stuck with 3D printers… haha.
Update: New HV PSU did absolutely nothing to solve the problem. Blech. We’re putting the full enclosure on it tomorrow, so we’re hoping that it’ll block out the radiated EMI, and if we leave the control board on the outside maybe it’ll work. Also ordering a Smoothieboard at this point & hoping that it’ll magically fix things.
EDIT: +Peter van der Walt quick question. As far as the Cohesion3D vs. Smoothieboard goes, it looks like the big advantage the SB might have is the Ethernet (and the theoretically more robust communication it might bring). However, it looks like this still isn’t fully implemented in either Laserweb3 or LW4. Is this correct? I tried to make sure that I RTFM on this one, but I’m not sure I fully understand.
@Alexander_Flick it might be better to continue the noise problem discussion over in the K40 Laser Engraving community this likely is not a LW problem.
I hope I am wrong, but my guess is that the new smoothie will not solve your problem. That is unless it somehow has better noise protection.
You must have some kind of high voltage leak and we need to find the source. Everything else will be a band aid, that is not to say I have never used a band aid before.
When HV like this arcs or pulses it can contain a lot of energy and it lifts grounds or droops power supplies in a short high energy pulse that resets processors. I have seen scope traces that have ionization rise times in ns for this tube.
We need to get on a hunt for the source … thinking of how… get back on this later after coffee.
