Power Supply Fault - TubeMic

[Shawnobi]

Bit of a long shot but figure I'll give it a go. The power supply for my Se 2200T tube mic decided to start smoking. I replaced a bulged 3300uF cap; started it up, and after 5 minutes the same component went again.
I can't seem to get a schematic, or a direct replacement from Se. And I really don't want to bin the whole microphone.
So before I start wildly stabbing in the dark and replacing most of the components, I wondered if any one may have some pointers/ suspects etc?
Fuses didn't blow and it was powered up on its own without mic attached. It switches on fine, goes into it's warm up phase where the LED blinks, then you hear the relay click in. Then about 5 minutes later that cap starts to fizzle and bulge!

Shot in the dark but appreciate any ideas. Thanks

 

[radardoug]

One of the diode bridges is probably faulty.

 

[rock soderstrom]

One of the diode bridges is probably faulty.

This is a possible source of error, also check your voltage selector switch if it is in the right position. (120/230V? )

What I noticed, 16 V rating for the smoothing capacitor of the 12V DC heater is not sufficiently dimensioned. The transformer says 11V AC for the heater before rectifying. That means 11V x 1.41= 15.51V DC raw at 3300uF cap and that is in loaded condition. Without microphone, i.e. without load, this voltage will be a bit higher. I would use at least a 25V type!

After testing the rectifier, you can check what voltage is applied to the 3300uF capacitor when the microphone is not connected.

 

[Shawnobi]

Thank you both, I've ordered some 2W06's and I'll grab a higher rated cap. I'll feedback as soon as I have the parts delivered and installed.
Much appreciated.

 

[chilidawg]

Use your multimeter to check the surrounding parts just to make sure they are ok. The last thing you want is after you put in the new parts is to have them blown up again.

 

[Goblin]

Use your multimeter to check the surrounding parts just to make sure they are ok. The last thing you want is after you put in the new parts is to have them blown up again.

It is good practice to keep one hand behind your back when working as well, or sit on it, to avoid shock.

 

[CJ]

Solder wires to cap that is bulging. Clip DC meter leads to cap. Plug in for a few seconds and measure voltage. Check DC volts first. Then switch meter to AC volts and repeat test.

Does DC voltage exceed cap voltage? Is polarity correc

How much AC voltage?
If like 10 volts AC then you have a problem with ripple current tearing cap up. Excess ripple indicates bad rectifier as mentioned above. Report back immediately . Don't worry about shock. Can't be worse than driving inferior Russian surplus tank past angry Ukrainian with javelin.

 

[mad.ax]

The big question is WHY a 16V capacitor?
Its proximity from the diode bridge suggest it's the first filtering cap.
Please take a pic of the PCB trace side in order to confirm this.

There are 3 secondaries on the transformer. 217V is obviously for the B+. Then we have 11 and 16 Volts AC. Could be 11V for the timer and logic supply, and 16V for the heater regulated down to 12... Or could be 16V for the logic (seems high and need some regulation for the NE555) then 11V for the heater which could be regulated to 6V...
Please measure the ac voltage on the two green wires so we get a definitive answer on this.

At any rate, be it 11 or 16V AC, that will be too much for a 16V cap, as Rock pointed out...
Are you sure that 16V cap hasn't been changed before?
What is the reference of the regulator with the heatsink?

Axel

 

[chilidawg]

Most likely the cap was changed before by someone without electronics background, because other than it looks physically different than the rest of the caps on the board, there's just no way the factory would install a 16V cap for filtering DC voltage rectified from either 11V AC and 16V AC.

Made in China is cheap but certainly not stupid.

 

[Rob Flinn]

Yes. Ground yourself as well so you don't get shocked. That can cause immediate death.

Why would you do that ? It's dangerous and you are more likely to get a bad shock.

 

[rock soderstrom]

Why would you do that ? It's dangerous.

+1, its way better to keep one hand behind your back.

 

[rock soderstrom]

Please take a pic of the PCB trace side in order to confirm this.

Please measure the ac voltage on the two green wires so we get a definitive answer on this.

What is the reference of the regulator with the heatsink?

I think these are important questions. There are too many possibilities. Check this before you swap components.

Is polarity correct?

Good point, most PCB designer would try to keep the ground markings on one side.

 

[mad.ax]

I imagine OP checked the polarity first thing... And he said the cap swells after 5 minutes... Methinks an inverted 3300uF would BLOW way faster than this... Unless the fuse dies first that is...

Axel

 

[Shawnobi]

Just browsing the comments this morning. Thanks for the info.
Without getting too distracted

Safety: yep. I've tinkered with and built tube amps with up to 500VDC plate voltage so pretty hot on safety and discharging caps.

Re. The 16V cap. It left the factory with a 16V rated cap. The one you see in the photo is the replacement I put in. Different cap, but same capacitance and rating. I can't comment on why Se chose to use that value. I installed with the polarity correct.

 

[Shawnobi]

Ok, numbers time.

AC Volts on the 2 green wires = 12.7 VAC

Across Failing Cap (3300uF/16V) voltages: DC: 15.5 VDC/ 0.5VAC

Regulator In (L7806CV) 15.5V

 

[chilidawg]

Where does the cap actually sit? If it's post regulator, then obviously that explained why they put in a 16V cap.

EDIT: oops sorry, didn't see "Across Failing Cap (3300uF/16V) voltages: DC: 15.5 VDC/ 0.5VAC"

 

[rock soderstrom]

Ok, numbers time.

AC Volts on the 2 green wires = 12.7 VAC

Across Failing Cap (3300uF/16V) voltages: DC: 15.5 VDC/ 0.5VAC

Regulator In (L7806CV) 15.5V

Is this with or without microfon?

Edit: added schematic, poor design IMHO

Edit2: cap polarity is correct.

 

[mad.ax]

If that thing really left the factory with a 16V cap, I can only imagine that it was originally designed with a 8 or 9V AC secondary transformer.
At some point they decided to use a 11V transformer instead, and some stupid beancounter decided that the original 16V cap could whistand 15V5, hence do the job without reworking the PCB...
Poor choice indeed. Using a 25V cap should solve the issue.

Axel

 

[Shawnobi]

@rock soderstrom - Measurements taken without Mic connected. The missing cap value from your sketch is 6800uF 16V

@mad.ax Whilst that may be a poor choice of cap, I can't believe it the reason for the fault. This unit has been working since 2009 and that value was in every unit on sale. Every unit would have burst into smoke after the first switch on. So Whilst I will upgrade to 25V, that surely won't fix it (perhaps mask it for a while longer). Something else must be going on. The new cap I popped in fizzled out less than 5 minutes after power on.

 

[mad.ax]

11V AC, aka 15,5V DC, into a 7806 is a poor choice. (why waste 10V when 5 would be enough?)

15,5V DC into a 16V rated cap is a poor choice. Might work for a while. Might even last after the warranty period. But it will overheat, and will fail someday...

Not neccessarily the designer's fault, could be a bean counter decision.

As for something else going on, I hope not! 0,5V AC seems to indicate that your diode bridge is good. Try the 25V rated elco. Check that your filament gets proper regulated 6V DC. Check that nothing overheat after say 30 minutes. If all this is OK, I bet it will work for at least another 10 years...

Axel