Can A PT Fail Providing Too High Voltage?

[Daver]

Absolute worst case this amp was fed from an overwound PT, so it had higher voltage from Day 1 , most factories had more than one transformer supplier , there might be batch to batch variations or they might have tried something different and even on a Friday afternoon they might have fitted a transformer meant for another amp but which was handy, just to finish the d*mn amp and deliver it so as to fulfill an order or get some needed cash.
All these things quite normal in the real world production life.

Personal example: my meat and potatoes amp is the 2 x 12" SS 100W one, I donīt care much (read that as not at all) for bedroom rockers and such, all my amps are meant for live use with a loud drummer nearby .
That said, many demand something lighter and more compact , so sometimes I agree to make a single 12" 60W one which I consider the absolute minimum for Club work.
Power amp board is the same, amp is loaded with 8 ohms (1 speaker) instead of 4 ohms (2 speakers) , and uses a somewhat smaller and lighter PT.
*Many* times I have to custom wind a single 60W one for the odd order because thereīs none left and I simply grab a 100W one instead from the shelf just to avoid the bother; in such cases a puzzled Tech may find "higher than normal voltages" ....... now you know why.
Same might have happened here.

I knew going in that it might just be the way it is, but I wanted to make sure I wasn't missing something before I do anything else like trying to reduce the voltage. Thanks.

[Mick Bailey]

I've never come across a failure mode where a PT produces a higher voltage. By that I'm thinking that one day instead of giving X voltage it now gives X+n. For this to happen intrinsically and with no other external effect or change, the turns ratio would need to alter. Not a possibility. The effective turns ratio could be increased by the primary winding having a good number of turns shorted out, but this also shorts out the magnetic circuit and kills the transformer - so no output (well, maybe a very small output but with massive current draw which would at least blow the mains fuse).

There could be an instantaneous situation where the transformer is arcing internally which could give rise to a flyback voltage being generated. This would likely be very short-lived and damaging. I have encountered a bias voltage being way higher than it should and this was caused by leakage from the HT winding. I recall this was external to the transformer though. I suppose one other possibility with a transformer could be an issue with primary voltage taps if it were a multiple-voltage primary. Thinking here that there would need to be some unusual connection/disconnection issue with the wiring or selector. Not an issue with a 'straight' transformer.

I see a lot of amps where the voltages are much higher than I would like. A cross-check is to measure the heater AC voltage and see how close it is to 6.3v. If this is high as well then probably the mains is higher than the transformer was specified for.

[bob p]

something else to consider - 6.3VAC filament windings are nominal. the actual tube specification is typically 6.3V +/- 10%. And if you look at a Fender schematic, they typically say that all voltages have a 20% tolerance. there's a lot of room for variability without there being a problem.

there have been a number of threads where people buy "replacement" transformers for a specific classic amp, only to find that the idle voltages turn out to be higher than expected. sometimes the transformer designers take pride in overbuilding their PT to assure that it will meet or exceed original specs. when they exceed them, it's not uncommon to see voltages that are higher than on the schematic. we're used to seeing overvoltage conditions when we change suppliers for replacement iron. it's just as reasonable to expect that the original builders had the same problems when they changed suppliers.

none of those voltage variations bother me, unless the heaters get beyond +10%. that will shorten tube life.

[Daver]

I just wanted to make sure there wasn't something I was missing. Looks like it's just a case of new line voltage vs. old. For those who asked, the AC out of the PT under load is 391-0-391 and the heaters are at 6.6vac, so everything is a bit high. Thanks!

[Mick Bailey]

6.6v isn't too bad. Some modern Fenders have 7.4v to 7.6v on the heaters here in the UK and often they've been running like that for 25 years with the original preamp tubes and sometimes the original output tubes, too.

[pdf64]

True, though that may speak more for the toughness of those tubes, rather than the advisability of exceeding the heater rating.
Some of those Fenders have a 260Vac mains tap which can help to get things more reasonable, though with others the highest available is 240V, and back to back diodes or low value metal clad resistors fastened to the chassis have to be used to get the heater voltage reasonable.

[Mick Bailey]

The problem with most Fenders is they're configured for "UK mains 230v". Despite our mains being 240v (and more). Annoyingly they usually have an 'Australian market' 240v tap (Devilles/HRDLX series etc) that just needs two wires swapping over to make a happy amp. Even worse, some amps bought direct from Europe are wired for 220v and remain so.

[pdf64]

something else to consider - 6.3VAC filament windings are nominal. the actual tube specification is typically 6.3V +/- 10%. And if you look at a Fender schematic, they typically say that all voltages have a 20% tolerance. there's a lot of room for variability without there being a problem...

I've seen some tube info that has a +/-5% limits on the 6.3V, though can't find it at the moment.
But really, given a mains voltage tolerance to nominal of ~5%, even for tubes whose heater tolerance is 10%, the PT heater secondary needs to be within 5% of 6.3 (and 5V if applicable) at mains nominal. And it would seem natural to specify the HT winding voltage 'tolerance to intended nominal' similarly.
So I don't think that the note regarding the 'voltage values shown + or -20%' should be taken as a simplistic factual statement; yes, that is what the words say but it simply isn't credible for it apply to the raw voltages directly from the PT secondaries, eg that one AB165 could pass QC with a HT at 425V, the next at 340V and the next at 510V.
Whereas it does seem reasonable that the voltages 'downstream', where normal tube variance and resistor tolerance will have their effect on current draw, hence HT node and early stage plate voltage, may easily vary 20% either side of typical. eg V1 pin 1 typical 220V, acceptable range 176- 264.

A note detailing what tolerances apply to which typical voltages noted on the schematic would get kinda clunky.