Danbridge 15kV insulation tester JP15 1993 repair teardown

Рет қаралды 1,102

Күн бұрын

Пікірлер: 13

@FireballXL55 Ай бұрын

Nice find. Those SG3524 IC's are getting harder to source these days. They are used in some of the kit I have to repair.

@tpa6120a2dwp Ай бұрын

Thank you for showing the inside of one of these. While YT is full of teardowns of all kinds of test equipment, such more specialized devices that not every lab or company uses are practically nonexistent. If you by any chance come across a combination wave transient generator, please make a video about it - as far as I know that would be the only one on the entire platform. When I design a new product, I occasionally need to do transient tests according to IEC 61000-4-5 with the 1,25/50-8/20µs transients. In order to be able to do this without each time paying hundreds to rent such a machine for a few days, I built a simple circuit according to the description in the standard. It is built around a soviet 6µF / 4kV capacitor and when it is discharged, it turns components into fireworks with ease (if there is no adequate input protection varistor in the circuit or the input fusible resistor is not capable of absorbing the transient). I used a homemade triggered spark gap as a switch, which works but is not trivial to get working at lower voltage levels. I'd love to see how commercial units do the switching.

@TeardownOZ2CPU Ай бұрын

i made a surge, load dump simulator unit for DC rail testing automotive many years ago, motor driven vario, step up transformer, caps, current limit serie sesistor, IGBT, voltage and current feedback, and pc interface and software for it all, we used it for many years, you can scale this idea to any IEC norm you need, and it is not that complex or expensive to make.

@tpa6120a2dwp Ай бұрын

@@TeardownOZ2CPU Thank you - I looked into using IGBTs but was not sure regarding the higher transient levels. Basically the norm calls for 500V, 1kV and 2kV. Unfortunately I design mainly industrial electronics, so I need the 2kV level at several kA. This means I need at least 2 IGBTs in series to split the 2kV to more manageable 1kV per device. I think I can build a gate driver that is powerful and triggers the gates practically simultaneously (did something like that for a different project years ago). Maybe if I build two banks of 10+ parallel strong TO247 IGBTs that are connected in series, the manufacturing variations of the individual devices should level out the more are paralleled, so the difference in turn-on time for the two banks should be small. Do you think this is a good approach? So far my experiences with voltages above 1kV and semiconductors have been - lets just say "expensive", that's why I went the stone-age route with the spark gap....

@TeardownOZ2CPU Ай бұрын

@@tpa6120a2dwp working with very high voltage and very high pulse currents, is not going to be a cheap hobby project, in parts and time invested.. sparkgabs can be connected in series, very common methode for large scale lightning tests, and they can be remote triggered, I would recommend IGBT more accurate and repeatable, Littlefuse make really nice IGBT parts 4500V 90A brand new to be released part number : IXYX40N450HV, but if 4500V 60A is enough then this one IXYH30N450HV is cheap and available. I can not imagine how you be able to make it with IGBT in series, voltage sharing and pulse shape matching = impossible.

@tpa6120a2dwp Ай бұрын

@@TeardownOZ2CPU Holy cow, I guess its been a few years since I last researched such devices on Mouser et al. - I was not aware of these. Thank you! If I aim for a max. pulse current of 2kA@2kV, using 10x IXYX40N450HV in parallel with a maximum pulsed collector current Icm of 350A per device should in theory give me up to 3.5kA maximum pulse collector current. From your experience, is going up to 200A Icm per device safe for a device that is specified with 350A if one would like to stay safe and not blow up 570€ worth of IGBTs? You are absolutely right regarding putting IGBTs in a series arrangement - I tried this for switching 5.5kV@10A and blew up so many parts and - worse - test equipment, it was not funny anymore at all. I ended up using a pulse modulator tetrode GMI83 from the USSR. It needs a lot of power to heat the gigantic cathode but it is right at home at its task, development only took a few days once I had thrown in the towel on the semiconductor version...

@TeardownOZ2CPU Ай бұрын

@@tpa6120a2dwp if you really want to know for sure the safe max current pulse, i test for this, using my voltage, my pulse, my drive implementations, only one device needed to get this wisdom, if it survice, then scale up, you are right : tubes are still used today ! and are still available for very high voltage and current pulse it is the only way to go, unless you go spark gab

@migsvensurfing6310 Ай бұрын

Nice battery charger 😄

@MIMALECKIPL Ай бұрын

problem is current contemporary batteries are not that resistant

@Duracellmumus Ай бұрын

Never use read relay as an safety switch, it can be actuating by an external magnetic field and a small ones respond realy fast by current pulses. The instrument is olso seems to be shielded against static fields only.