2500 Watt ZVS Induction Heater excessive current control protection update (Part 2)
A few months back I added a PID temperature controller so I could turn the current on and off as needed to maintain a programmable temperature or average power level, using a DC to DC SSR. After testing several inexpensive DC to DC SSRs, I found a couple that seem to work well as long as they NEVER see a serious spike in current. When they do, they short, permanently. After realizing that simple fuses can't protect the SSRs, I realized that I needed something much more sensitive. So now, this is Part #2 of building a fast acting overcurrent sensing device that can shut off the SSR's trigger pin BEFORE the thing fails. Here, I show my progress to date. I now have a device that is almost ready to be installed. That will happen in Part #3.
More at: www.spaco.org
Пікірлер: 13
I have had ssimiliar problems when using only a single channel of a quad op-amp...the solution is simple: On all unused amps, connect it's "-" input to the output AND connect the "+" input to ground. ...PRICELESS VIDEOS!
At about timestamp 5:11 maybe that pair of sunglasses in the case had something to do with the problem LOL!
@dalenassar9152
Жыл бұрын
Seriously though, any unused op amps must each have the inverting input (-) connected to the output, and the non-inverting input (+) grounded.
Nice work chap! Love your films. Can you post the schematics and code? I had pretty much given up on the royer circuits as I could not make an effective over current protection and has a huge number of dead mosfets but maybe this could be the answer
@frenchcreekvalley
2 жыл бұрын
Thank you for your kind words. Regarding: "Can you post the schematics and code?: I will wait until I get the "SSR Protector" actually up and running (and, hopefully, doing what it is supposed to do) before I do that. Right now I have a lot more messy code in the UNO than is needed. Regarding "dead Mosfets" I haven't had any problem with blowing Mosfets for quite a while now. Even when I cause short bursts of current beyond the 50 amps that my analog meter reads, the Mosfets don't blow. If you are having that sort of problem. see this playlist: kzread.info/head/PLVxWen9M87dBhvInCQ-3pELWdxCM3XH4I I do a LOT of explaining and troubleshooting in many of those videos. Particularly, see: kzread.info/dash/bejne/a351r9p_j5u-ktY.html if you are short on time.
excellent sir, i need to know ; i can connect two zvs 1.800w and parallel in one coil
@frenchcreekvalley
2 жыл бұрын
I do not know the answer to your question. I suggest that you buy one of the 4000 watt units that are now for sale if you have not already gotten the 1800 watt units.
What do you think of current controlled turn-off from simply monitoring the current supplied by the DC power supply input with a compactor?
@frenchcreekvalley
Жыл бұрын
Hello Dale. When you say "compactor", do you mean "contactor" as in "relay"? If so, I would be concerned that about the speed of a relay turning off fast enough to save the ZVS Mosfets. And,---- a relay rated at 50 amps DC will be quite expensive. You would also have to consider the circuitry needed to to drive the relay and the speed of that part of the system. I am not saying that it wouldn't work, but I avoid relays in high current DC circuits because of contact arcing.
@dalenassar9152
Жыл бұрын
@@frenchcreekvalley I totally agree with your reply about mechanical switches and stay away from them for exactly the reasons you stated. I also worry about "contact bounce" here. COMPARATOR is the word (solid state) I intended to write. It came up as misspelled (like it's doing now). Looks like it is not in spellcheck, and I overlooked the replaced word....sorry.
Nice work thank you.is it possible to make ZVS super fast, i mean resonant frequency about 500kHz.
@frenchcreekvalley
2 жыл бұрын
I don't think you can run them that fast. The faster you try to run the circuit, the longer percentage of the time the Mosfets are in linear mode, causing them to heat substantially. In my opinion, running them faster than about 130kHz is a recipe for board failure.
@newmail3549
2 жыл бұрын
@@frenchcreekvalley thanks for your reply.