mystery output transformer impedance

[alchemy]

I am working on an old homemade 2x6l6 amp that has old Motorola transformers (i'm guessing an old radio or something. The secondary wires are black, green, and green/white. The green and green/white are hooked up to the speaker output (16 ohm).I put it on a variac and green and green/white I get a turns ratio of 15.8:1 and an impedance ratio of 250:1. Multiplied by a 16 ohm output I get about 4000ohms, not too far off for a pair of 6l6's.
The thing that is confusing me is that when I put voltage across the primary and measure the secondary from either of the greens to black, I get half of the above value and both sides are equal. I have never seen a transformer wired like this (or maybe just never bothered to notice). Usually the black (mostly) is common and across black and the other wires are different impedances. This one seems to be divided exactly in half.
So my question is: Does it matter which way it's hooked up, other than matching it correctly to the output tubes?

[MWJB]

In plenty of amps where the black is supposed to be common...it's not. Transformers often don't match expectation or legends, I always double check with a voltage test.

If the the green and green/white taps give the highest voltage, then one is the common the other is the 16ohm tap.

If the amp has a NFB loop, then polarity of the secondaries might be an issue. The amp will howl if it doesn't like primary/seconday polarity wiring.

[alchemy]

In plenty of amps where the black is supposed to be common...it's not. Transformers often don't match expectation or legends, I always double check with a voltage test.

If the the green and green/white taps give the highest voltage, then one is the common the other is the 16ohm tap.

If the amp has a NFB loop, then polarity of the secondaries might be an issue. The amp will howl if it doesn't like primary/seconday polarity wiring.

Yes that make sense. The part thats confusing me is that if one of those IS the common, then it means that the black wire is tapped exactly at the center because the voltage is identical from blk to green and from blk to grn/wht. I've never seen that before. If that is the case, then does it matter (when wiring for 8 ohms) which way it's hooked up(grn to black or grn/wht to blk)?

[MWJB]

If the black to grn/white voltage is exactly half what you get from green to green/white, it seems unlikely that you will also get half voltage from black to green. Half the secondary voltage equates to a 31.6:1 turns ratio and a 4ohm tap (31.6x31.6 = 998ohms)?

For an 8ohm tap you should be seeing 0.7VAC at the secondary, with 15.8VAC at the primary.

Are you sure that there isn't a 4th wire (8ohm tap), perhaps hooked up to a NFB tap?

Also make sure that when you measure the voltage output at one side of the transformer, that you monitor the voltage input at the other side, when you select different taps, the input voltage will fluctuate, I always use 2 meters simultaneously.

Otherwise, the taps could be 15 & 7.5ohms...and the different impedance shown to the meter is enough to corrupt your 0.68VAC that you should be reading. Don't necessarily assume that the taps are 16/8, or that you are specifically looking for 4K on the dot.

[alchemy]

If the black to grn/white voltage is exactly half what you get from green to green/white, it seems unlikely that you will also get half voltage from black to green. Half the secondary voltage equates to a 31.6:1 turns ratio and a 4ohm tap (31.6x31.6 = 998ohms)?

For an 8ohm tap you should be seeing 0.7VAC at the secondary, with 15.8VAC at the primary.


Are you sure that there isn't a 4th wire (8ohm tap), perhaps hooked up to a NFB tap?

Also make sure that when you measure the voltage output at one side of the transformer, that you monitor the voltage input at the other side, when you select different taps, the input voltage will fluctuate, I always use 2 meters simultaneously.

Otherwise, the taps could be 15 & 7.5ohms...and the different impedance shown to the meter is enough to corrupt your 0.68VAC that you should be reading. Don't necessarily assume that the taps are 16/8, or that you are specifically looking for 4K on the dot.

That's exactly what I am saying, that the blk to grn voltage is exactly half of the green to grn/wht, and that the blk to grn/wht is also exactly half. I put a 14.3vac signal to the primary, measured .91vac across grn to grn/wht. Across the other two cmbinations I measured .45vac. I was using two meters, monitering the primary voltage as I went. There is not another wire on the secondary. I wasn't looking for 4k, That's just what I got.

The whole reason I was checking it in the first place is that the amp sounds like it could have a bad OT. Fresh output tubes and a speaker swap had no effect. While going through it looking for other issues I decided to do a quick check to make sure that the windings were not open on one side and also (because it is a pretty jenky homemade 5c8 copy) check that the OT whoever built it used had the correct impedance.

So you can see why I'm stumped. I would not expect the other secondary conections to measure exactly the same, nor both exactly half of the other. And If I wanted to hook it up for 2k-8ohm, would it matter which side I used?

Or is there something else going on? Turn to turn short? as

[Steve Conner]

A center tap on a 16 ohm winding is a 4 ohm tap, relative to either end. (This weird situation comes about because impedance is proportional to the square of the number of turns.)

I think the RETMA code specified green for the speaker ground, black for the first tap, green with stripes for the second one and so on. I remember having trouble with this in my Crown restoration, trying to figure out which green was ground and which was the 8 ohm tap.

I've seen home-made/modified amps where the OT taps that the constructor didn't want were carefully rolled up in electrical tape and tucked inside the transformer end bell. (that was where my 16 ohm taps went) If you seem to be missing a tap, it's worth taking the bell off for a look.

If you suspect a bad OT, RG Keen has some articles on testing transformers at geofex.com. Or you can just apply 120V to the primary. (A good OT will stand this easily, but any shorted turns and it will go up in smoke)

[MWJB]

I'd also try applying a low voltage (0.5 - 5VAC?) to the secondary, then seeing if there is a mismatch in voltages from either end of the primary to the centre tap, to identify any shorted primary windings.

Not because I doubt Steve's assertion, just because you mention a degredation in tone, I'd just want to rule it out.

[alchemy]

[QUOTE=Steve Conner;208643]A center tap on a 16 ohm winding is a 4 ohm tap, relative to either end. (This weird situation comes about because impedance is proportional to the square of the number of turns.)

I think the RETMA code specified green for the speaker ground, black for the first tap, green with stripes for the second one and so on. I remember having trouble with this in my Crown restoration, trying to figure out which green was ground and which was the 8 ohm tap.


Yes, it would be 4 ohms to a 4k primary (I meant 8k for 8 ohms). That's how it is wired, green is the speaker ground.

I'll definately put voltage across both sides of the primary.. Not sure how that slipped my mind. The resistance was very close between sides, so I guess I just assumed it was OK.