eccerr0r
Solder Sloth
Last Activity: 09-03-2026, 12:01 PM
Joined: 11-08-2012
Location: CO
-
kind of hard to do if this is a real 2n3055.. where'd you get it and who made it? -
Seems to be just being driven off mains! There doesn't really seem to be anything else...Leave a comment:
-
The gate drive of the dead IGBT (low side) is fairly simple it seems, mostly just a resistor and the zener. A regular diode too but I replaced that to no avail. The other high side driver is much more complex but that IGBT never failed, adding to the mystery.
Another annoying thing about these failures is I lose a fuse along with the MOSFET/IGBT each attempt. I guess I got lucky getting 10 fuses for $1 once but running out of these fuses. When I end up having to use fuses at $1/pop is going to start sucking if I can't get it to last a year per pop.Leave a comment:
-
Well only one of the three blows oddly enough, and the same one always blows despite being symmetric on the other side of the H.
The other three transistors are not shorted at all, still seem to be functional. I suppose if there are any suspects it'd be the other one on the totem and punch through would be the most damaging... and that transistor seems fine. Not sure what the exact mechanism is causing the failing transistor to die, especially because it still works sometimes. for a while. or whatever... Really need to probe the circuit I suppose but need isolation...
None...Leave a comment:
-
what's troublesome is that it worked for quite a while just fine, but now it fries a specific transistor every few days or so...
circuit looks identical and matches to the other half bridge but that half bridge is fine/no blown transistors.Leave a comment:
-
Yeah probably that one leg has something wrong with it, it blew again. The faster turnoff of the mosfet wasn't enough to save it from self destructing.
It's a discrete circuit and will be a pain to reverse...
Right now I've got nothing except possibly replacing the gate protection zener diode but usually those don't fail open? Not to mention gate drive appears to be current limited by a 1KΩ resistor...both up and down.... and because of this I don't see too much back damage that can be caused by shorted DGS (or CGE).
Otherwise shitty inverter design...
And...Leave a comment:
-
Unfortunately it still blew. A few times. Once it blew right as I turned it on. Ugh. Might be a circuit design fault or other issue with circuit? Hmm.
Now I put in a 18N50 back instead of using the 80N60 and 60N100s as an experiment. Currently running a 18N50, two 60N100s, and a 80N60 in the H-tree: one underpowered TO-220 MOSFET and three TO-264/TO-247 IGBTs...
So far it's working today. After being out of service for a month I lost several dollars worth of electricity *sigh*. Need every kWh I can get, especially after now needing a J1772 connector...Leave a comment:
-
PCs will sometimes spike high in current when some background task runs so you can't assume idle all the time... yeah running an intricate benchmark will set it to highest utilization and your UPS needs to be able to handle it whether or not it's a "real" task...Leave a comment:
-
yay to personal password hints in low traffic public forums!?!Leave a comment:
-
The intent was for a custom AC watt hour meter so I'll also need to deal with the same issue with the CT. I suspect most people cheat with using a resistive direct sample as it's cheaper and less device variability that could be a problem with any sort of isolation transformer, but since this is a one-off project I don't care too much about cost since I already have the transformer... but isolation is the key issue.Leave a comment:
-
Heisen is correct (approximately, of course this depends on the actual transistor used). Depending on the ability of the pot, the discrete 139 transistor will probably get more current through than the 675a darlington in your configuration.
Only ways to get more current is to increase your voltage, use common emitter (PNP if needed), or decrease your load resistance;, otherwise you're topology limited.Leave a comment:
-
I suspect bipolar junction transistors can short out one of the two junctions and have the other one still work like a diode, so it's possible universal testers may identify it as a diode.
Still suspicious, usually these tiny to92 transistors even when shorted won't cause power fuses to burn...thing would have to pass quite a bit of current to do so and they weren't designed to do so...Leave a comment:
-
I bumped up my PSU to 10A and looks like we have a live one, tripped pretty much right away, so looks good.Leave a comment:
-
Ok since it was damaged due to physical interaction, you're probably fine with either one.
I don't know about the policy of this forum anymore. Before when it was run by TC I would most definitely suggest uploading. Now with the change in command and policy change, I now have a change of heart.Leave a comment:
-
What is the device and circuit that's using these 1n4148's? I would hope that the circuit is not running up against the thermal limit making these ratings meaningful, these diodes were meant to be low current, non-critical applications...Leave a comment:
-
I wonder if I hit it with a charged capacitor would it instantly trip? Heh. Unfortunately don't know how much current it would be passing...Leave a comment:
-
usually resistors go open when they fry so it should be fairly quick to just see if any semiconductors shorted...running these custom boards out of the original device is quite a challenge...Leave a comment:
-
Testing a circuit breaker...
I suppose people should test their parts before deploying...or should they?
I got a questionable "button reset" 1962-vintage 5A circuit breaker from a junk pile. I was wondering it still worked or was it welded closed or something.
So I tried to pass current through it using my SCR adjustable PSU. Seems to pass 5A just fine...
Then I hit 6A... still no reaction...hmm.
Waited a minute...still no reaction. Fearing damaging my PSU (or wasting electricity...then again as a SCR adjustable PSU it should be fairly efficient even at 6A) I... -
Phase shift while using an AC line transformer...
Well, ideally I measure the voltage straight from line voltage but anyone tested how much phase shift adding a line frequency transformer adds? At radio frequency, transformers clearly add phase shift, but at 50 / 60 Hz, how much of one could I expect, assuming one doesn't connect it such that I get 180° phase shift...
Curious if it's still legitimate to measure instantaneous line voltage by measuring output of a transformer and multiplying by turn ratio. But if there's a significant phase shift, this wont work for me, but it would be nice to get some sort of isolation... -
Not clear what you're trying to accomplish here... normally I don't use a breadboard and just test while the device is installed, or is it because it's in an inaccessible location?...Leave a comment:
No activity results to display
Leave a comment: