PowerTech MP3098 20A 13.8V linear power supply repair

[VK3YNV]

Here is a weird but interesting power supply circuit. I have two of them to repair, both with the exact same symptoms.

This exact same model is made with literally dozens of different brand names and comes in a variety of current capacities. from 10Amps to 50Amps.

There is no over voltage protection, and an unusual over current protection system. The over current relies on the voltage drop being detected and
tripping V30 to turn off, when that happens current flows through R3 and V29 to the base of V31 and shuts off the output driver. The funny thing is that it will stay in this state until you power down completely to reset the latch.

There remains a problem however, exactly how does the thing start up in the first place and turn on V30? There is a diode V19 that feeds a 3.6V zener V20 via a 220R resistor and generates a pulse on power up that couples through C8 that turns on the base of V30 via diode V21, once V30 is turned on the output stage can start to regulate.

So the nett effect is a short delay on power up and maybe that stops the initial inrush current from overloading the array of 2N3055's ??? dunno

Downsides to the design are... running diodes in parallel is generally bad design practice, as is running 2N3055's without emitter resistors, but the
real killer for me is the lack of overvoltage protection. So I will be fitting crowbar protection.

The fault symptoms were the output voltage was dropping with very light loads, a small current load would drop the output voltage to just a few volts. Same fault on both power supplies the fault turned out to be a bad transistor V28 had a bad base emitter junction. Replaced the 3DG9014 with a 2N3904 ( same pinout higher current). And now both are working again.

Circuit below comes from jaycar's web site.

Here is a video, by Jordan VK3ACU fitting crowbar protection to the MP3098

 

[BillC]

Here is a weird but interesting power supply circuit. I have two of them to repair, both with the exact same symptoms.

This exact same model is made with literally dozens of different brand names and comes in a variety of current capacities. from 10Amps to 50Amps.

There is no over voltage protection, and an unusual over current protection system. The over current relies on the voltage drop being detected and
tripping V30 to turn off, when that happens current flows through R3 and V29 to the base of V31 and shuts off the output driver. The funny thing is that it will stay in this state until you power down completely to reset the latch.

There remains a problem however, exactly how does the thing start up in the first place and turn on V30? There is a diode V19 that feeds a 3.6V zener V20 via a 220R resistor and generates a pulse on power up that couples through C8 that turns on the base of V30 via diode V21, once V30 is turned on the output stage can start to regulate.

So the nett effect is a short delay on power up and maybe that stops the initial inrush current from overloading the array of 2N3055's ??? dunno

Downsides to the design are... running diodes in parallel is generally bad design practice, as is running 2N3055's without emitter resistors, but the
real killer for me is the lack of overvoltage protection. So I will be fitting crowbar protection.

The fault symptoms were the output voltage was dropping with very light loads, a small current load would drop the output voltage to just a few volts. Same fault on both power supplies the fault turned out to be a bad transistor V28 had a bad base emitter junction. Replaced the 3DG9014 with a 2N3904 ( same pinout higher current). And now both are working again.

Circuit below comes from jaycar's web site.

View attachment 3433

Here is a video, by Jordan VK3ACU fitting crowbar protection to the MP3098

Could C7 be a critical suspect there Ray?

 

[VK3YNV]

Hi Bill,

Yes, I thought C7 might be a likely suspect. But turned out to be V28 base emitter junction has gone open.

I have already ordered the VK3ACU crowbar pcb from JLCPCB, and got 5 boards for the grand total of $US3.60 including shipping

 

[VK3EB]

My Powertech supply picked up from last meeting was switched on only to have smoke and a very nasty smell and frying noise fill the shack. Strange as the unit was in perfect condition and looked like it was never used, DC out was 13.8, but I switched off quickly, unit had no load.
Needless to say I wasn’t expecting that result, next, take the lid off see what’s going on. Nothing obvious burning, I had my suspicions it maybe main filter electros, frying noise was the clue, so this time listen very carefully as I pulse the power switch on and off. more frying noises which stopped after few second, next checked the 4, 4700uF/35v with my fingers which we hot. Bingo, problem found. Removed and replaced with new electors rated at 50v. The unit was in new condition so why crook caps, I think the supply was unused for years and the sudden surge of voltage at switch on the rated 35v caps broke down, rail is about 26v.
Nice little project replacing caps and fitting VK3ACU overvoltage protection board with front panel fuse holder. I fail to understand why design a power supply and not have DC fuse protection, it does have a 4amp AC fuse but that’s useless with things burning up well before it blows. The other thing I add was a LED to show volts on the output terminals.

 

[VK3YNV]

Nice neat job with the crowbar, and I like the fuse on the front panel.

Reforming capacitors is a thing that's usually required after really long periods of being powered off, typically many years, particularily in VFD drives, the trick is to power them up at low voltage and gradually increase the voltage over time while watching the current. Then finally test the ESR.

Funnily enough there was a huge heavy box in that shed, that was a capacitor reforming set. I think Graeme took it down to Geoff Angus's shed.

 

[BillC]

An old bloke said to me one time fuse every thing . I,m sure he was right. Good save Dallas

 

[vk3vsm]

.

Funnily enough there was a huge heavy box in that shed, that was a capacitor reforming set. I think Graeme took it down to Geoff Angus's shed.

no i didn't grab that , it was a heavy thing, it's still in the shed

 

[VK3EB]

Good lesson on powering up old gear, I have reformed caps in the past mainly on valve gear with high voltage ratings (600v etc) I wasn’t expecting a problem on 35v not so old caps.