Hypersensing questions

snowgoer540 wrote:
Why not 24K 10W?

I trust PCW. This project was his baby!

snowgoer540 wrote: Why not 24K 10W?

No need.

rodw wrote:
I trust PCW. This project was his baby!

Me too, and i have been doing electronics since i was a bloody kid, but when PCW speaks, i listen very carefully.
Hell, sometimes i get paid to destroy stuff on purpose, so i do a lot of testing with everything that i can get my mitts on, except when i work on industrial stuff, no experimenting there.
PCW has a different way of doing things, he has to make things work for years and years without any issue, so just thinking about them takes much longer, let alone implementing.

tommylight wrote:

snowgoer540 wrote: Why not 24K 10W?

No need.

Plus he has calculated 3.75 watts and rounded up to 5 so there is enough headroom. (>30%)

Its actually good a couple of water tables got on the bandwagon as I did the initial testing without a water tray..
Also great we have a couple of brands of plasma cutters involved too.

With a bit of luck, Ill have some time tomorrow to work on this.

rodw wrote:

snowgoer540 wrote:
Why not 24K 10W?

I trust PCW. This project was his baby!

Oh, no doubt. I didn't mean to come off as if I was questioning him, just more of I wanted to understand the math/reasoning behind it for my curiosity.

Did I miss where he calculated the 3.75W? Or is there another thread I am not seeing??

EDIT: I see it in his post for recommended value now. I apologize.

So, I want to put the hypersensing circuit on a relay as @PCW suggested. Is it better to kill the main power, or would it be better to cut the 24V DC side? It seems like it would be faster to just switch the DC side, and leave the Power supply powered up all of the time. Any thoughts here?

Also, I'll likely need some help adding the ohmic enable output pin to the hal file.

No matter how much documentation I read, wrapping my head around the hal file connections doesnt seem to stick.

snowgoer540 wrote: Ok good! At least we have a solid baseline.

I havent had time today to do any work on the plasma, I was going to use a small wattage resistor as you did here, and not fire the torch. But I think that saves me some testing. Is there a reason you used 22K ohm over the suggested 24K by Peter?

For the hal scope tests you are doing, can you describe exactly what you are doing to test it? I want to make sure I can perform the same tests to reliably test what's going on. I am glad that the resistor takes the sensitivity to a wet torch down. That's a giant step in the right direction in my opinion.

I am still wondering... is that the use case Rod is trying to fix via code, or is he testing something else?

Used a 22KOhm over 24K just because that's what I had. Shouldn't make a big difference except in wattage requirement if you stray too far.

Power = V^2/R so for a 24K resistor at 300V, you get 3.75W required, 4.09W for a 22K

With hal scope I'm running 2 channels. Ohmic-volts on one channel and ohmic-on with the other. Set the trigger based off of ohmic-on and let it capture the data. Or I just keep it on "Rolling" and manually stop the data capture so I can get data before and after the event.

Thank you for the maths! How would one settle on 24K over 10K?

My guess would be that the trade-off is high sensitivity= lower sense current - more susceptible
to leakage from water in torch vs low sensitivity = higher sense current - worse at detecting contact
with dirty surfaces, though this may be improved with a higher sense voltage.

Sorry to sort of but in, but I have been reading through this forum and then many others tonight and there seems to be a lot of negative press around ohmic sensing and water tables. It got me thinking about a whole other way of non contact sensing with say a proximity sensor or some other device completely left of centre. In my research I came across this technology by swift cut. They claim it’s like a feather sense that can sense even the thinnest material and I’d say for 99.99% of cases it would work flawlessly. Could this be a solution or even a technology added into Plasma C to take it to the next level?

Attached is the brochure of the plate sensor and a demo Video , I couldn’t find a good view of it cutting thin material but have seen on Instagram it sensing 1.2mm with no sweat.

If this should have its own topic created let me know and I apologise in advance but I thought most people who would have an interest in this are replying to or viewing this topic.