Transfomer failures

[Randall]

I know we always say here that transformers are the least likely suspect in amp failures, but... Of the seven amps in my shop right now, four of them have bad transformers. Four out of seven! A vintage Ampeg, a Dynaco stereo block, a Blackface Pro reverb, and a Blackface Deluxe Reverb. I'm not 100% on the last one, but I've run up against the wall so many times on this freak , I'm throwing an OT at it to see what happens. Dark days indeed.

[Jon Snell]

Mains transformers fail due to incorrect fuses on overload or bad intermittent mains supply. Mains transformers draw a lot of current at switch on if the mains is above the zero point and get warm but if continuous poor/dirty mains is supplied may incur shorted turns and develop more current losses until they destroy themselves.
Output transformers usually survive a lot of abuse but can arc over if seriously over driven with no loading or if the wrong HT fuse is fitted not protecting the primary windings from over current causing heat and that could cook it causing insulation to break down and produce short circuits. The latter usually gives very weak sound with no bass response before failing completely, sometimes going open circuit.
As you say, very unlikely under normal circumstances, a lot of work just to test like that. I induce a 10volts sweep from 150hZ up to 10khZ into the secondary 8 Ohm tapping, and measure the primaries with no valves plugged in. It is soon apparent if there is an issue and you will soon get to know a good transformer from a bad one by comparison.
Measure the input current and look at the primary waveform with a 10k Ohm resistor as a load on each side.
Saves an awful lot of time!

[nickb]

In my experience I've seen more bad transformers then you might expect from the 'last thing to suspect' adage. Worst offenders are toroids where I often see a shorted turn near the center where the wires overlay each other.

Transformers can fail in many different ways so any test has to consider each of them. With big iron, shorted turns, open windings and insulation failures pretty much cover it, In ferrites you can get cracked cores too which can be hard to spot.

The switch on surge can certainly be huge, especially in toroids, But it is quite short lived so little heat is generated and does not usually cause a problem. Incidentally, the current is highest when the power line waveform crosses zero since the transformer has considerable inductance and volts and current are 90 degrees out of phase in an inductor. It's complicated as the core can saturate for a short period while the core 'charges' and the current surge depends on the actual phase at switch on and the remnant magnetic field in the transformer.

Insulation can be damaged by abrasion during winding or mishandling, dielectric breakdown or heat. So power line surges, asymmetric power waveforms and overloading are causes.

If you have a shorted turn the primary current will be very high since you are driving a short circuit. Easy to spot using a bulb limiter and open secondaries (disconnect or isolate as necessary) for a power transformer. For an output transformer you can feed the secondary from a heater winding via a 10 ohm resistor ( take care with common grounds if you opt to use the amp's own heater supply) and check the current by measuring the ACV across the resistor. It's there mainly to protect your driving transformer. If high it's bad. Beware high voltages on OPT primary when testing. Another very easy method is to 'ring' the transformer using a battery a neon ( search for details).

Open is rather obvious. You either get no primary current or no secondary voltage. Use a load on the secondary as you can get fooled if it's not a high impedance open, especially at higher voltages.

Inter-winding insulation is tricky. You really want a hi-pot tester of some kind and check the inter winding and winding to core resistance using high voltage DC.

[Helmholtz]

I think that apart from normal aging/corrosion processes, increased line voltage (especially in the US) contributes to increased transformer failures rates in vintage amps. Higher amp voltages mean higher components' and total amp dissipation and thus higher internal temperatures. Generally higher temperature increases failure rates.

[Rhodesplyr]

The historical fact is that some manufacturers, even back in the golden age of tube electronics, kept manufacturing costs down by using transformers that were running close to their limits in their applications. I've been working on Dynaco gear for decades now, and Dynaco amps do suffer PT failures. Ampegs are also known for PT failures. I have a replacement upgrade PT by Triode Electronics in my Dynaco ST-70, and it runs MUCH cooler than the original. It's also quite a big larger with a taller lamination stack.

It's also the case that these things are now often over 50 years old, and I'm pretty sure that the manufacturers did not have that kind of lifespan in mind when they built them.

I work on a lot of Hammond Organ equipment, and they started being built in 1935. We see a lot of winding failures in pre-WWII electronics. However, in all the many Leslie amps I've worked on, I think I've replaced two OTs and no PTs I can remember.

I've got a Silvertone 1484 on my bench right now, and its OT was replaced back in the 1980s, judging by the work I find.

[Chuck H]

All the above rings true, of course. And I don't think the "transformers are the last thing to suspect" adage applies to what comes over the counter at repair shops. Many problems are solved by users, so they never make it into your shop. And a great deal of transformers are replaced by novices trying to fix an amp when the transformer was never the problem. I think that statement applies to the novice DIY community at large. How many posts have we seen where where a novice writes (paraphrasing) "My amp quit working. Is the transformer bad?" Anything that ends up in your shop will be a repair the user couldn't manage on their own. So, a higher ratio of blown transformers.

So it's still a practical phrase here on the forum. Repair techs usually know how to diagnose symptoms and when to check the transformers.

[SoulFetish]

Incidentally, the current is highest when the power line waveform crosses zero since the transformer has considerable inductance and volts and current are 90 degrees out of phase in an inductor. It's complicated as the core can saturate for a short period while the core 'charges' and the current surge depends on the actual phase at switch on and the remnant magnetic field in the transformer.

exactly right.

TE has a great app note Beware of Zero-Crossover Switching of Transformers. It's brief, and does a great job illustrating this. (Did for me anyway)
https://www.te.com/commerce/Document...=CS&DocLang=EN