Ampeg B-15N Power Transformer Failure

[mjrippe]

This is more of an informative musing post than a problem that I need help with, but input is welcome!  Schematic attached.

I had a vintage Ampeg B-15N on my bench today that was "blowing fuses" according to the owner.  I had worked on this amp previously and knew it was working fine within the last year.  However (there's always a "however"), it had been recapped by someone else prior to that.  They used good caps, but WAAAAY oversized the first filter stage with a pair of 250v 560ufd in series!  They also added diodes across the tube rectifier from pins 4 and 6 to pin 8.  Two 1N4007 in series from 4 to 8 and two from 6 to 8.

So I open the amp up and find a broken 5v filament wire from pin 2.  "Easy fix!" I think, as I re-solder the wire.  Pop in a new fuse, flip on the Variac and just touch the knob - it's drawing a full amp at about 5 volts input.  Crap.  Shut it down, pull all the tubes (which I had already tested BTW), and try again.  Same deal.  Start snooping around with my bench meter.  Primary looks good, filaments ok, get to the B+ and each leg measures ~30 ohms to the center tap BUT from Red wire to Red wire I'm getting about 0.5 ohms.  Check the diodes and they *seem* ok, but then I de-soldered them just to be safe.  Still under an ohm.  De-solder the B+ wires from the circuit completely and yup, still shorted.

So obviously the winding is shot, but here is a question - would this have happened without the diodes and huge filter caps?  It seems to me that the rectifier tube would warm up slowly enough that the inrush current of the large caps would not overtax the B+ winding (correct me if I am wrong).  Having the silicon diodes in place allowed those caps to charge fast and pull a lot of current.  Of course then it wouldn't matter if the tube was there at all, right?  So why would the broken filament wire matter, if it did?  Odd one fore sure, but if I rebuild it I'm yanking the diodes and oversized caps!

 

[Audio1Man]

It is hard to say what the real problem is as your data is not complete.

I would  suspect that the 1N4007 diodes may be shorted.  I would remove both of the red wires  and them make measurements and turn on the AC power and see if the current is high.

1. Is the standby switch in STANDBY MODE?
2. Check resistance of each red wire to ground.
3. Is the tube socket shorted?
4. Remove the RED wire from the tube socket.

Another trick is to power the device from full AC Supply with an light bulb in series and then check voltages inside the device,
The lamp can range from 25 to 150 watts as needed.

Good Hunting. Duke

 

[CJ]

excess C failure mode wouldprobably be too much wire current = open winding,not shorted,

those old transformers used varnished wire which gets old to the point where you can scrape it off with your finger nail,
so an upgrade would add depedabillity,

then you can worry about the OPT,

 

[panman]

It just rings in my ears, that the first filter-cap after the rectifier-tube should never exceed 50u.  Well, that`s about it then.

 

[mjrippe]

It is hard to say what the real problem is as your data is not complete.

I would  suspect that the 1N4007 diodes may be shorted.  I would remove both of the red wires  and them make measurements and turn on the AC power and see if the current is high.

1. Is the standby switch in STANDBY MODE?
2. Check resistance of each red wire to ground.
3. Is the tube socket shorted?
4. Remove the RED wire from the tube socket.

Another trick is to power the device from full AC Supply with an light bulb in series and then check voltages inside the device,
The lamp can range from 25 to 150 watts as needed.

Good Hunting. Duke

Yes, I tested in Standby mode and On.  I not only removed the diodes from the circuit, but also removed the Red wires from the PCB (and tube socket) completely.  Red wires to Ground obviously depend on setting of standby switch but are equal to Red wires to Red/White center tap.  I will quadruple-check that it is still drawing crazy current with them disconnected tomorrow.  Thanks!

 

[mjrippe]

excess C failure mode wouldprobably be too much wire current = open winding,not shorted,

those old transformers used varnished wire which gets old to the point where you can scrape it off with your finger nail,
so an upgrade would add depedabillity,

then you can worry about the OPT,

Very good point CJ - thanks for the input.  Though excess current may have caused an insulation failure as opposed to burnt out winding?  Not that it matters, a new power transformer is needed either way :-(

 

[mjrippe]

It just rings in my ears, that the first filter-cap after the rectifier-tube should never exceed 50u.  Well, that`s about it then.

Yup.  Regret not changing it out the first time I saw this amp.  There's always this stupid part of my brain that says "Oh, maybe that guy had a good reason for doing it that way and I'm not smart enough to know why!"

 

[CJ]

i could rewind it for you, that way you do not have to drill new holes in the vintage amp and try to shoe horn an aftermarket product,

also could drop turns on sec a bit  as those were designed for 110 so you get a lot of failures in old amps due to tubes blowing up from excess B+, groupdiy would also dig the data,

cj

 

[lassoharp]

So I open the amp up and find a broken 5v filament wire from pin 2.  "Easy fix!" I think, as I re-solder the wire.

Makes me wonder if the 5AR4 wasn't already shorted.  Or maybe it shorted when you re-connected the filament

They also added diodes across the tube rectifier from pins 4 and 6 to pin 8.  Two 1N4007 in series from 4 to 8 and two from 6 to 8.

I found this response from another forum.


"Adding a silicon diode in series with the plate of the tube rectifier increases the peak inverse voltage capacity over just just that of the 5AR4. Doing this allows you to put a larger capacitor after the tube rectifier. Without it, you can get flash over in the tube when you first power the amp up due to the current draw of the charging capacitor. It serves to protect the tube rectifier.

As an example, Fender does this on their 57 Deluxe reissue amp http://support.fender.com/schematics/guitar_amplifiers/57_Deluxe_Fender_57_schematic.pdf."



My gut says that there is fine line being walked on here (being at border of preventing tube arcing).  Sounds like something could easily not work as expected- tube arcs and shorts and takes out the PT Pri . . . . .  The diodes are supposed to protect the primary, yet you have a dead short at 500+ VDC sitting right on top of the primary.    Maybe combo of high PIV on the insulation and the PT simply being old and having many hours ran at high temps conspired at the right time and killed it.

I've melted PTs to where the varnish comes spewing out of the sides and not actually caused an open or short.  It went into a purgatory - could only make about 1/4 of its normal output voltage.  No idea what this meant inside but I suppose a number of things could happen

 

[mjrippe]

i could rewind it for you, that way you do not have to drill new holes in the vintage amp and try to shoe horn an aftermarket product,

also could drop turns on sec a bit  as those were designed for 110 so you get a lot of failures in old amps due to tubes blowing up from excess B+, groupdiy would also dig the data,

cj

Hi CJ, repro potted power transformers are available for around $225 that should be drop-in replacements.  If you think you can do the job for less please PM me.  It would be interesting to see the dissection

 

[mjrippe]

lassoharp - I did check the 5AR4 (and all other tubes) on my TV-7 tester before trying the amp.  Interesting info about the diodes, thanks for that.  I knew I had seen it done before but didn't quite understand why.  As I mentioned, when the tube was *not* rectifying due to no heater voltage, perhaps the diodes took over and let the caps draw too much too fast?

In any case, I checked again this morning with the B+ secondary completely disconnected and it is clearly shorted.  Amp starts pulling current immediately with no tubes, no diodes, no nothing connected!  "She's dead, Jim".

 

[lassoharp]

perhaps the diodes took over and let the caps draw too much too fast?

Yeah, everything getting slammed harder with the SS diodes only in there.  May have just been the last kicks to an already dying horse.

 

[Biasrocks]

Fliptops offer a period correct power transformer, not cheap but it looks the part.

http://www.fliptops.net/catalog/p-100561/power-transformer-pt108-for-ampeg-b15nb12n

Mark

 

[mjrippe]

Fliptops offer a period correct power transformer, not cheap but it looks the part.

http://www.fliptops.net/catalog/p-100561/power-transformer-pt108-for-ampeg-b15nb12n

Mark

Hi Mark, Yes - that is the one I was referring to.  I was $4 off on the price though ;-)

 

[CJ]

you could by the flip tops xfmr and send me the old one to dissect just for kicks,

i can help you with shipping $$

probably about 6 bucks a piece for medium flat rate box

 

[mjrippe]

you could by the flip tops xfmr and send me the old one to dissect just for kicks,

i can help you with shipping $$

probably about 6 bucks a piece for medium flat rate box

Excellent suggestion.  I have to talk to the owner of the B15 but he might be fine with that.

 

[andyfromdenver]

Did you remove all connections from the HV secondary, including the bias feed, and get an AC reading?  Is there a burned smell? 
Beyond that, assuming it needs replacing, you should decide whether you want to run rectification Solid state or Tube and rebuild the amp accordingly.  The other tech who put 280mfd as the Pwr supply filter was likely, inteding to run it solid state hence the diodes and perhaps disconnected the 5v winding on purpose, none of that sounds good imo anyway.  Before you totally chuck it.  test the PT with voltage and the secondary completely disconected and you can assertain beyond a doubt if the PT is shot. 
  Of course if the guy has the $ an upgrade as CJ recommended is always good on a vintage amp if it is loved etc.