Odd OT Impedances?

[zendragon63]

Every time I think I kinda know something, it usually turns out that I don't know it quite well enough. I have had a output transformer in the heap that came out of a Asian built 18 watt and while I ASSUMED that the marks were correct, I measured it to be more sure. What I got puzzled me a bit; it is like the output taps may not have got tapped in the right place--see attached.

I checked my approach and math a couple of times:

vac primary/vac secondary=vac ratio. Vac ratio^= impedance ratio. Impedance ratio * speaker impedance= reflected impedance.

I have several known multi-tap transformers that appear to calculate out pretty close to what they are marked so it seems like the these 4 and 8 ohm taps on this OT may have been taken at an incorrect point in the winding process.

That said,

1) Does the calculations on my attachment look correct and
2) any problem just using the 8 to 16 ohm taps (yield a 9331 Ra-a)? Any insights are welcome and thanks in advance. Regards

dennis

[DummyLoad]

please review attached schematic. i think you have taps of the secondaries in reverse order. seems to be around 7.6-8.0K Ra-a OT. 3.2ohm is not common nowadays, you don't mention the age, so still not certain.

--pete

[zendragon63]

Thanks dummy, I can't open the .sch files here at work so I will check it out when I get in tonight.

This is a 7 or 8 year old OT from a Ceriatone 18W and the secondaries are plainly marked 'Com-4-8-16' on the side of the OT right where each of the wires are. I checked this several times because it seemed so odd. But numbers are numbers.

This OT was pulled and replaced with a Heyboer (though surprising enough, the improvement was minimal) so it is just looking for work.  Again thanks! Regards

dennis

[zendragon63]

Right you are dummy. It is indeed reversed--can't say why it left the factory that way but it is what it is. (Mental note: don't always trust the labeling on them out-of-town winders.)

As for the OT, it's future full-time employment now looking much brighter.....   Thanks again! Regards.

dennis

[DummyLoad]

yw! happy rosin smokin'

--pete

[DummyLoad]

so shouldn't the measurements be from the center tap to either primary lead?  

not necessarily. ratings on a PP OT are anode to anode (Ra-a). you can measure from CT to either end, then you simply double the results.

--pete

[HotBluePlates]

The voltages do not look right.  I always thought O/T's were step down transformers.

They are. But unless you have 300-400vac available to swing the entire primary, you generally apply a voltage to the secondary and measure the result on the primary.

Additionally, because they are step-down devices, you have potentially greater accuracy/precision applying a voltage to the secondary and measuring the result on the primary. 120vac applied to the primary of a 8kΩ:8Ω OT gives you 3.79vac on the secondary; same voltage on a 4kΩ:8Ω OT gives 5.37vac on the secondary. If you think you need exact numbers, the step-up from secondary to primary makes big impedances changes translate to big (and easily measurable/calculable) voltage differences.

[PRR]

> Attached is a link on testing Output Transformer Performance.

Working Link:
http://www.dalitech.com/Resources/Measuring%20Output%20Transformer%20Performance.pdf

[rob_h]

The simple way I look at transformer taps is that the voltage from the 4 ohm tap is from "x"turns of the secondary winding, producing "Vx" volts due to the turns ratio of primary inductivly coupled to that portion of the secondary winding.  The 8 ohm tap should be taken from "2x"-turns producing "2Vx" volts.  The 16 ohm tap should be taken from "4x"-turns and would produce "4Vx" volts.

With a perfect transformer, yes, the voltages induced into the secondary taps should be multiples of the 4 ohm tap: .110v, .220v, and .440v.  Even though they are not perfect multiples, They do track in the right direction, but with losses.  .195/.220=88.6/% and  .354/.440=80.45%, so it looks to me that there is nothing backwards here, just not perfect, in my humble simple view.

I might put a 1 ohm resistor in series with the primary, hook up 4, 8, and 16 ohm loads, and compare the voltage drop across the resistor to effectively measure the primary current.  For me, the more information I have, the more pieces of the puzzle I have.  It's a puzzle! Good hunting!

rob_h

[PRR]

> 4 ohm tap is from "x"turns of the secondary winding, producing "Vx" volts due to the turns ratio of primary inductivly coupled to that portion of the secondary winding.  The 8 ohm tap should be taken from "2x"-turns producing "2Vx" volts.  The 16 ohm tap should be taken from "4x"-turns and would produce "4Vx" volts.
> but with losses.

But that thinking, extended, leads to "wrong" answers.

And it's not "losses".

Remember: Power is Voltage squared divider by Resistance

0.110v in 4 ohms is 3mW
0.220v in 8 ohms is 6mW
0.440v in 16 ohms is 12mW

However we know that the amp should deliver the *same* power into each of these impedances.

So that's wrong.

The right answers would be:

0.110v in 4 ohms is 3mW
0.155v in 8 ohms is 3mW
0.220v in 16 ohms is 3mW

So for impedances of 1 : 2 : 4 the turn-ratios are 1 : 1.4 : 2.