DIY impedance matcher?

[birt]

can i use an old 220V-110V step down transfo as an impedance matcher?


i was thinking.. it has a 1:2 ratio. so if i have an 8 ohm output and a 4 ohm cab, and i connect the output of the amp to the "2" winding and the cab to the "1" winding, does the amp "see" 8 ohms?

it would be a nice thing to have on the bench...

[sluckey]

The impedance ratio is the square of the turns ratio, which is also the square of the voltage ratio.

So, your 1:2 voltage ratio would be a 1:4 impedance ratio. It's likely that the wire used in that transformer is too small for a low impedance/high current device.

[birt]

thanks :)

[sluckey]

I may be wrong about that!  :-[

Better check on Aiken's site, or hopefully someone will clear it up.

[PRR]

> was thinking.. it has a 1:2 ratio

1:2 Voltage, 2:1 Current, Impedance is Voltage divided by current.

What Sluckey said.

> It's likely that the wire used in that transformer

Yeah.

For very low loss, the planned impedances would be similar to the original impedances.

If the 240V winding were intended for 16 ohm load, that would be 240V/16R= 15 Amperes, and 240V times 15 Amps is 3,600 Watts. So it seems a 3,600 Watt 240:120V could be used as a 16:4 transformer.

You surely don't have a 3,600W tranny, or want to lug such a lump around,

A 120V 120 Watts winding is clearly 1 Ampere, and the design load was 120 ohms. A ~~100VA tranny probably has 10% loss, much of this in the copper, which implies it has about 12 ohms of copper resistance. If you used it to feed a 4 ohm speaker, at-best you would get 25% power in speaker and 75% power in winding heat.

You could do the same with a 12R resistor between your 16R tap and your 4R speaker. And a helluvalot lighter.

You should check distortion. As a very rough guide, the audio AC should be less than a third of the rated power AC. "120V" winding, keep audio down to 40V or it will distort. (Power transformers DO distort in operation, but the crap reflect back to the big utility generator, not our teeny tubes.)

You should consider high frequency response. PT only claims 50Hz or 60Hz. In fact it "must work" to 400Hz so cap-input rectification works good, and it costs nothing to give that to you. If the primary and secondary are simply wound one over the other, 5KHz is not uncommon. If they are wound on separate bobbins, you may not pass 1KHz well.

> it would be a nice thing to have on the bench...

Find a good deal on a high-power 4/8/16 OT. It can have a blown-open or ripped-out primary, as long as the secondaries are good.

[birt]

thanks for the ideas, i'll try to find an old OT to use. if i have a 4-8-16 ohm OT's secondary i can make any conversion right?

[HotBluePlates]

if i have a 4-8-16 ohm OT's secondary i can make any conversion right?

If I follow PRR right, then yeah.

Here's what I *think* he means:
You could attach your 16 ohm output from the test amp to the 16 ohm taps on your impedance matcher. Next, let's say you have a 4 ohm speaker load you want to attach to the test amp. Simply hook this load to the 4 ohm taps on the impedance matcher.

Meaning you don't hook up the primary at all. This is what I think PRR meant by saying your transformer could have a "blown open or ripped out primary."

This would make the impedance matcher into something like an auto-transformer, basically meaning that a certain amount of volts-per-turn is induced in the secondary by your attaching the test amp to a selected tap. When you select another tap, you keep the same volts-per-turn, but change the number of secondary turns, meaning you change the resulting voltage to the speaker.

Play with the equation for power using volts and resistance/impedance: Power = Voltage² / Resistance

Let's say the power output is 40w. You take this output power from a transformer having 4, 8 and 16 ohm taps.
Voltage = SqRt (40w * 16 ohms) = 25.3v
           = SqRt (40w * 8 ohms)   = 17.89v
           = SqRt (40w * 4 ohms)   = 12.65v

You can see that you still get 40w, even with the lowering voltage because with a lower impedance, more current flows. You can work through ohm's law and the formula for power to prove that each of these voltages results in 40w.

I'm kinda kicking myself that I didn't remember this earlier...

Oh... and your multi-tap output transformers provide different output impedances with the same power through this exact same action.

[birt]

i actually understand everything said in this topic. doesn't happen that often ;-)