REDD EQ, Helios 69 and Dick Swettenham

[ruffrecords]

Thanks Ian! Yes, I did see that, but I wasn’t sure what to make of it. So I should add a 10k resistor to ground — RL on the schematic above — off of the yellow lead between the transformer and the eq input? (Understanding that the zobel network is then unnecessary).

Apologies if this should be implicit and obvious from your answer.

No you are right it is not implicit. What it means is that if you can arrange for the load attached the the transformer secondary to be 10K then you do not need the Zobel network. So if your load is 20K for example then you could add 20K across the transformer to make its load equal to 10K.

Cheers

Ian

 

[JMan]

Thanks, Ian!

I still don't fully understand how I will approach this from a practical perspective, but it seems like I need to go do some research about understanding load impedance. I've been googling a bit and so far everything I've found on transformer loading is way over my head (and not audio related, though I imagine many concepts remain constant). I'll try and wrap my mind around this before coming back with more questions!

 

[ruffrecords]

Thanks Ian! Yes, I did see that, but I wasn’t sure what to make of it. So I should add a 10k resistor to ground — RL on the schematic above — off of the yellow lead between the transformer and the eq input? (Understanding that the zobel network is then unnecessary).

Apologies if this should be implicit and obvious from your answer.

It is difficult to answer questions helpfully when you don't know how knowledgeable is the person asking. You have to make assumptions and often they are wrong. This is not your fault it is just life. So first things first. Do you understand ohms law?

Cheers

Ian

 

[JMan]

It is difficult to answer questions helpfully when you don't know how knowledgeable is the person asking. You have to make assumptions and often they are wrong. This is not your fault it is just life. So first things first. Do you understand ohms law?

Cheers

Ian

Hey Ian,

I totally get it! You're nothing but helpful, and it's truly appreciated. I sure hope I haven't given you any other impression. I'm a hobbyist with no background whatsoever in science and engineering, so "explain it like I'm five" is probably where I am with things. But I can promise that I do my absolute best to try and understand what people tell me.

Yes, I understand ohms law as a principle and as an equation (I like the triangle that folks use to visualize it, that was very helpful to me when I learned about it). However, I lack experience with actually using it in the practical mathematics of a real circuit, and specifically I find that I am unsure of how to derive usable numbers for the variables in practice. Most examples I have seen assume that you already know two of the three and go from there.

 

[ruffrecords]

OK, so next question, do you know how to work out the combined resistance of two resistors in parallel?

Cheers

Ian

 

[JMan]

Yes, that one I understand!

 

[ruffrecords]

Yes, that one I understand!

OK so if you looks at the schematic in post #419 you will see there is a string of series resistors from the input to 0V. If you add those all up (because they are in series) you get about 27K. If you recall from earlier I said the EQ is designed to be driven from a source impedance of 2K5 so we add a 2K4 resistor in series with the input (I will explain why later) so this gives us a total of 29K4. The transformer we know prefers a load of 10K so we need to work out what resistor to put in parallel with 29K4 so that together they make 10K. The answer is 15K (to make the numbers easy call 29K4 as 30K. So 30 x 15 is 450 and 450 divided by 30 + 15 is 450/45 which is 10K).

So now the transformer is happy but what about the 2K4 source impedance for the EQ. Transformers do what they say on the tin - they transformer voltages, currents and impedances. They have no intrinsic impedance of their own. So, although the Jensen transformer is labelled 10K:10K that only means that, due to the practical limitations of making real world transformers, it works best at that impedance over the audio band. However, you will probably drive it from an interface of some sort which will have a relatively low output impedance - maybe 100 to 200 ohms or maybe even less.. The transformer will transformer this so the source impedance of the secondary will appear to be that of the interface i.e 100 ohms or so. But we want the EQ to be fed from about 2K5 which is why we added a series resistor of 2K4.

In practice it is even more complicated than this because the transformer is not perfect and it windings will also have resistance and in a 10K:10K transformer these often amount to a few thousand ohms. So if we connected the EQ to this transformer we probably do not need the 2K4 series resistor. This means our original calculation for the parallel resistor is wrong; it really needs to be about 16K but 15K is close enough.

Bottom line, stick a 15K across the transformer secondary and feed it straight into the EQ.

As you can see, audio design is not a trivial exercise.

Cheers

Ian

 

[JMan]

Wow, Ian, thank you for that explanation. This is incredibly clear. I really appreciate your patience and willingness to talk this through with me.

And for the record, if you are saying audio design is no trivial thing, imagine what it must look like to me!