Simple transformer question

[skrasms]

Despite having gone through a class that was supposed to cover everything we needed to know about transformers (which apparently doesn't include impedance ratios), I still have no idea how to actually USE a transformer in a circuit. Anyway, as I look at the schematic for the AM-864 compressor, the input transformer is shown as 600:10k ohms, and the output is 10k:600 ohms. Does that mean that I could use the same transformer in both places, just hooked up reversed for one side? According to my father there would be "power loss and attenuation" using a transformer in the direction other than it specs at, but I don't really know. Jensen has a 10k:600 ohm line out transformer, and it made me wonder if I could use it in both spots.

Sorry if I posted this question in the wrong area... it was sort of project-specific and sort of a general transformer question, so I wasn't sure where to take it.

 

[CJ]

For line in and line out applications, you should be able to get away with it.

There might be shielding issues, as the inputs sometimes have internal shields.

Now for mic in and line out, you probably would have to treat each case individually, depending on the iron used.

 

[edanderson]

the general and short answer, is that yes, the same transformer would probably work. but because transformers aren't ideal devices (nothing is), it will work better one way than the other. a 4:1 "line out" will work as a 1:4 "line in" but you may get some signal loss using it backwards.

specific to this application, the output transformer for the am-864 supplies the power to the output circuit via a center tap in the primary. this means that either you need an output transformer that can tolerate DC current or one where the two windings that make up the primary are matched for DC resistance, so that the current will be balanced. the jensen doesn't say if it is wound this way. to experiment, i would recommend buying two inexpensive edcor 10k:600 WSM series transformers. if they don't work for this, you'll find another use for them, and they're less than $9 each.

ed

 

[skrasms]

What is the significance of having the DC pass through the input and output transformers? Is the center tap through the in/out actually doing anything? I thought maybe it could just be ignored. Is it that the output is measured with the center tap as ground? Sorry if this is very obvious. This compressor will be my first project going completely from scratch armed with only a schematic.

Thanks for the clarification so far, guys!

 

[CJ]

If a transformer that is not designed to handle dc is subject to dc, the inductance drops and you will lose signal, along withn some other gremlins, like excessive heat, fine windings burning up on open circuiting (is that a word?), distortion and permanent magnetization of the core steel, which must be reversed in order to get the transformer back to normal.

Transformers that are designed for dc usually have a gap in between the lamination halves, which disrupts the magnetic and electrical path.

And the steel will be different. 80 percent Ni is usually too permeable for dc work, you may find some 50/50 used with a little dc, and some grain oriented M6 used for heavy dc in an output circuit.

 

[skrasms]

I understand why DC passing through a transformer not designed for it can be bad, but why do the input/output transformers in the AM-864 need DC to pass through them? That's what I meant to ask.

If I just didn't hook up that center tap the output stage would still be getting the same power coming from the 5Y3, wouldn't it?

 

[CJ]

Center tap of the power transformert? You need that to run the 5Y3 properly.
Do you have a schematic we can llok at? It makes life a lot easier.

Hold on. You are going to be my power supply Meta 101 guiena pig.

Scanning docs.....

Try not to electrocute yourself until I get back in ten minutes.

 

[skrasms]

Nono, not the power transformer center tap, that one makes sense to me.
I'm talking the one on the audio output/input transformers. I'll get a schematic...

Here we go: http://www.waltzingbear.com/Schematics/Misc/AM-864.htm

 

[CJ]

Oh, that transformer. Think of the center tap of that transformer as just a regular ground of a single phase transformer.
Then think of it as a ground for another signal that is 180 degrees out of phase with the first transformer. That's all it is.

The signal from both of those tube plates has to have a return path to ground, which is the center tap.


Darn, I thought I had a pwr supply study guide guinea pig. :sad:

 

[skrasms]

Ohhhh, I see, so the input amplifier is one where it both the positive and negative swings each get their own gain stage, right? I know we learned what that was called in my transistor class... darn summer-break memory lapses

In that case is it still considered that DC is passing through the transformer (and possibly causing the aforementioned problems), or just providing a reference? Basically I'm asking if that's a standard feature on line input/output transformers or if I'll need to look for a special kind and check like edanderson said.

And how does a normal XLR/TRS (or TS) wire into that?

Thanks for taking the time to answer all this for me!

 

[CJ]

In your circuit, d-c flows thru both pri winding halves at the same time, but since the d-c is entering the transformer thru the center-tap, the two magnetic fields induced in the core from the d-c are actually opposite in sign. Thus, the effects of the d-c tend to cancel, as long as the amount of d-c going thru both transformer coil halves is prety cloes to the same value.

The net effect of the "balanced" d-c is in theory, zero, although certain discrepancies, such as different winding resistances between the centertap and each end of the windings or different amounts of turns in the two halves of the coil due to the winder taking a coffee break during the middle of the winding, etc, can lead to the core seeing a d-c bias.

Different resistance readings in the coil are most likely caused by the fact that as the coil is wound, the length of each turn increases slightly. This small effect can really add up if there are a lot of turns in the coil. Some x-former makers get rid of this problem by winding the transformer in a "balanced" fashion, which may be done with a bi-filar winding (two wires being wound side by side, each representing one half of the coil) or by using two coils on a U core instead of one coil on an EI core.

 

[skrasms]

Okay, things are making more sense to me now.

I looked at the wiring of the Jensen and it's a no-go. Hammond has transformers that would work, but they're only spec'd down to 300hz. Is that normal? I figured audio transformers would spec out a lot nicer than that.