I have two Heathkit A9 output transformers that I would like to use for a PP design. When I measure the resistance of the primaries, I see the following. Center tap to primary 1 = 108 ohms, center tap to primary 2 = 126 ohms. This is the same for both transformers. When I look at the schematic, it looks like they treat each side of the primary differently. On the 108 ohm side, they add a 500pf cap to ground. I've noticed a similar thing looking at a Fisher 100 amp schematic, where they list a different primary resistance for each side of the transformer right on the schematic.

I have to assume there is a design reason for making the transformer unbalanced like this. Can anyone explain why they do this? Until now, I would have thought that you would want each side of the trans to be identical.

Thanks,
Mark

I'm sure this is not the reason. But if you wanted the unit to be used in a multiple voltage environment, you would have several taps on the primary winding.

Here's my guess.

There is a difference between DC resistance and AC impedance. I think the turns ratio in the transformer is the important element here, and it takes less wire to make say, 200 turns on the inside of the tranny than 200 turns on the outside of the first winding. The 126 ohms you see is just a longer length of wire making the same amount of turns as the 108 ohm leg. Perhaps by specifying the different resistances, the designer insures that the transformers will be installed in an electrically identical manner? Or perhaps the DC resistance plays an important part in the circuit as well.

Again, this is just speculation on my part.

Sounds like the small cap is maybe there to ground out high frequency oscillations? If it goes to ground I don't see how it contributes to feedback..

Sounds like the small cap is maybe there to ground out high frequency oscillations? If it goes to ground I don't see how it contributes to feedback..

That is better than a guess. The A9's were fairly far down the list as far a output transformer quality goes. Years back an engineer associate of mine devised a frequency sweep test using an AP1. We tested ~50 or so output transformers including Marantz, UTC, Altec, Heathkit, etc. We had a single A9 output and I recall it being pretty horrible. Lots of undamped HF garbage above ~15kHz.

Thanks for the responses. I always suspected these were not the greatest transformers. I built an amp with them a while ago, and when I traced through the circuit I found that almost all of the distortion was coming from the transformer. My square waves were very un-square.

I think I found a "why" regarding the unequal primaries. In the Radiotron designer's handbook, 4th edition page 222, there is an explaination about how transformers are wound. Apparently, it is difficult to get good coupling between the primary and secondary, so several winding schemes were used. The one that fits the heath says the following.

"This is an arrangement which gives minimum primary capacitance, but suffers from the defect that leakage inductance and winding rsistance are unequal for the two primary halves. For Class A operation using valves requiring an optimum load of high impedance this arrangement is sometimes the best".

They tell why the transformer is like this, but they don't specifically say what to do with it in a circuit. I was going to use these to build a PP triode amp, so it may be usable for that, but I'm not sure if I have to give any special treatment to the unequal halves.

Mark

Thanks for the responses. I always suspected these were not the greatest transformers. I built an amp with them a while ago, and when I traced through the circuit I found that almost all of the distortion was coming from the transformer. My square waves were very un-square.

The nice thing about the A9's though is that they are a great foundation to build a better amplifier on. Unlike W4's & W5's from Heath, the A9's seem to never torch their power transformer. Plus, the output transformer footprint is universal to many potted outputs such as the Acrosound TO-300, Dynaco A-420, and so on.

You can also splice in nearly any Scott or Fisher 7868/7591 output and run 6L6GC's. The only other mod to consider is adding a choke to the power supply although that certainly isn't critical. Plenty of options if you decide to take another path.

Has anyone considered 110 and 240? Living in Australia, I normally find xfmrs set up for a choice of either two halves of the primary in parallel for 110, in series for 220, and a little extension tap on only one half gives a primary suitable for 240.

Hi Indm,

This is the output transformer, not the power transformer. I guess I should have been more specific in the first post.

Mark