Which input xformer (600:600, 10k:10k)?

Hi!

I'm really unclear about how to determine input and output transformers. I have two I could use for the input of my experimental opto comp (Kiiratronics Konvertable). I have a Triad A-61J 1+1:1+1/1+1 with 2 windings in and 4 out 600:600 all. And I have a Jensen JT-11P 10k:10k. If the balanced output from the preamp is 600 ohms it will still drive the 10k input ok right? I'd rather use the Triad 600:600 though because it is smaller and easier to mount. The input tube of the comp makeup gain is a 6sn7 .

BTW is there anything that the split secondary Triad could be good for except making a mic splitter? I'm just trying to think of stuff to make this compressor more interesting after I basically get it working. Ha I mean IF I basically get it working :roll:

One thing I was vaguely thinking of is something like a second circuit/amp for creating "color", distortion actually I guess, which could somehow be dialed in feeding back into the input of the first or second stage of the makeup gain amp. I have no idea how or if this would be doable but it would be a cool thing. Besides, I hate doing things halfway and if I'm going to screw this up I want it to be spectacular. :razz:

The makeup aamp is a 6sn7 gain stage into a 6bx7 CF or maybe WCF, I'm still deciding but buiting the turret board to be able to do anything with the circuit I want. Probably means spray painting it orange and putting it on th wall as conceptual art.

thanks!

Kiira

10K:10K and 600:600 are both 1:1 turns ratio, but 10K:10K usually implies a higher primary inductance, which means less low frequency loss when being driven from a higher source impedance.

You say the preamp output is "600 ohms", but that doesn't really tell me anything. It could mean that the source impedance is truly 600 ohms, or that it's lower than that and capable of working into a 600-ohm load.

Really, the best way would be to try both and measure the difference. Also, you may need to adjust the secondary termination for best results with either transformer. Do you have equipment to check frequency response, check square-wave response, etc.?

...it also depends what you're going to feed into it of course.

If you're going to plug an API with a musclebound, steroid-enhanced 'output-stage-that-looks-more-like-a-power-amplifier' into it, then a 600Ω input impedance will be no real challenge.

If however, you try to drive it with a TL072-which-always-struggled-even-before-I-Accidentally-plugged-it-into-the-lightswitch-last-week-and-now-it's-not-at-all-well' output stage and then ask it for some mighty signal level, the 10kΩ might well be the better option.

But -for instance- an LA2a has a 600Ω input impedance, and all of my gear is quite happy driving it to decent levels.

Keith

[quote author="NewYorkDave"]10K:10K and 600:600 are both 1:1 turns ratio, but 10K:10K usually implies a higher primary inductance, which means less low frequency loss when being driven from a higher source impedance.

You say the preamp output is "600 ohms", but that doesn't really tell me anything. It could mean that the source impedance is truly 600 ohms, or that it's lower than that and capable of working into a 600-ohm load.
[/quote]

Yes the mic pre I'm planning on building right now has an output of it looks like less than 200 ohms at least. That could change of course. I want to build all of this to be 600 ohm capable just for the heck of it and I may be using it with other equipment in other people's studios too. I'm also planning on making both Hi and Lo-Z inputs and outputs on everything ala the NYDave 2 bottle., which is working so great. (must do sound clip).

Really, the best way would be to try both and measure the difference. Also, you may need to adjust the secondary termination for best results with either transformer. Do you have equipment to check frequency response, check square-wave response, etc.?

Click to expand...

Ok I will build it so I can substitute either and see how it works. Everything else is getting built to be rearranged as well.

Yes I think can measure frequency response. I have a scope and a function generator with square wave and a signal generator and they all work. Would I need a frequency counter? I could borrow one from my best friend if I do. The Jensen seems to be terminated with its own Zobel.

So there is no rule or maths equation to figure out what ohms I need for my xformers, like figuring out which cathode resistor to use? All I know is that the stronger an output is, low ohms and high current, the lower ratio output xformer can be driven. (I think)

thanks so much Dave,

Kiira

Oh, there's math for damn near everything in electronics. It's just a question of whether or not you choose to use it. Sometimes it'll save you some work, sometimes it's quicker to do it empirically.

With the secondary unloaded, frequency response will be down 3dB at the point where source impedance equals the inductive reactance of the primary.

Xl = 2piFl

If your source impedance is truly 600 ohms,

L = Xl/2piF = 4.77H for 3dB down at 20Hz.

In practice, you'd want your 3dB-down point to fall an octave or more below the lowest frequency you need to pass... In the case of full-bandwidth audio, that would be 10Hz. So a 4.77H primary wouldn't really be adequate for a 600-ohm source impedance. 10H or more is what you'd typically find in a high-quality 600:600 transformer.

But you must also take into account that with the secondary loaded, the reflected load impedance appears in parallel with the primary inductance. Which means that in the case of a 4.77H primary inductance and a 600-ohm source impedance, the loss at 20Hz will be greater than 3dB, the exact amount depending on the secondary termination.

This is an easy first approximation that assumes a resistive source impedance and does not take parasitics into account.

So, you can see why higher primary inductance is generally better, at least up to the point at which the copper losses and capacitance from the extra turns start to become a problem.

[quote author="NewYorkDave"]Oh, there's math for damn near everything in electronics. It's just a question of whether or not you choose to use it. Sometimes it'll save you some work, sometimes it's quicker to do it empirically.

With the secondary unloaded, frequency response will be down 3dB at the point where source impedance equals the inductive reactance of the primary.
[/quote]

I see. That is so helpful to me because I can't "understand" things with math I just get numbers but if someone tells me the relationships between the parts of the circuit I can actually understand better.

In practice, you'd want your 3dB-down point to fall an octave or more below the lowest frequency you need to pass... In the case of full-bandwidth audio, that would be 10Hz. So a 4.77H primary wouldn't really be adequate for a 600-ohm source impedance. 10H or more is what you'd typically find in a high-quality 600:600 transformer.

This is an easy first approximation that assumes a resistive source impedance and does not take parasitics into account.

Click to expand...

Yay. I'm finally getting the idea much better. That's what I needed to know to begin to get the relationships I have to worry about somewhere that makes sense. I could never understand why inputs didn't usually have a higher impedence because I had the sort of kind of idea that lower impedance outputs were able to drive higher impedance inputs but not the other way around.

But you must also take into account that with the secondary loaded, the reflected load impedance appears in parallel with the primary inductance. Which means that in the case of a 4.77H primary inductance and a 600-ohm source impedance, the loss at 20Hz will be greater than 3dB, the exact amount depending on the secondary termination.

Click to expand...

So the termination of the input helps to adjust for various input inductance? Or impedance/inductance mismatches?

It seems like it sorts out to being a good thing to have the source impedence and the input be close to equal and when they are not equal then various things can be done to help like terminating the input transformer advantageously. I guess this is also the rational for selectable transformer taps or resistor which can be switched and why pads aren't simple series resitors they have to be made to keep the input impedance fairly constant.

This is an easy first approximation that assumes a resistive source impedance and does not take parasitics into account.

So, you can see why higher primary inductance is generally better, at least up to the point at which the copper losses and capacitance from the extra turns start to become a problem.

Click to expand...

If I have it right or close then you have totally explained something I've not been able to understand for a long time. I know how tubes work very basically, like how a grounded cathode amp works but I just haven't had much idea of how stages or source/inputs interact with each other, how to connect tubes together in other words. Thanks again.

I really would miss you if you left Dave, and not just because you are a good teacher. I read our post about CBGBs and it was moving, so I guess like the music we've made on a good night, it's gone in one way but not another.

Kiira