Mic preamp input impedance and transformer impedance

[Newmarket]

The impedance of an inductor varies with frequency getting bigger as the frequency gets lower.

???? Z(Inductor) is proportional to frequency (Z=jwL)

 

[ruffrecords]

???? Z(Inductor) is proportional to frequency (Z=jwL)

I  am an idiot. I'll change the other post.

Cheers

Ian

 

[Newmarket]

I  am an idiot. I'll change the other post.

Cheers

Ian

Easily done - proportional / inversely proportional / impedance / admittance ...

 

[ruffrecords]

Easily done - proportional / inversely proportional / impedance / admittance ...

Let's just call it a senior moment.

Cheers

Ian

 

[Newmarket]

Let's just call it a senior moment.

Cheers

Ian

It's just the reciprocal of youth  ;D

 

[saint gillis]

In this thread https://groupdiy.com/index.php?topic=6571.msg859723#msg859723 Pucho shares a measurement method. Basically apply a 1kHz voltage across the primary, put a resistor across the secondary that drops the signal by -3db, a resistor in series with the signal generator that drops the output signal by another -3db, the resistor values give us the primary and secondary impedances (of course there are errors due to the output impedance of the generator etc).

What do you think of this method, and then what these impedance values correspond to physically?

 

[ruffrecords]

In this thread https://groupdiy.com/index.php?topic=6571.msg859723#msg859723 Pucho shares a measurement method. Basically apply a 1kHz voltage across the primary, put a resistor across the secondary that drops the signal by -3db, a resistor in series with the signal generator that drops the output signal by another -3db, the resistor values give us the primary and secondary impedances (of course there are errors due to the output impedance of the generator etc).

What do you think of this method, and then what these impedance values correspond to physically?

It is OK as far as it goes. What it measures is the impedance of the primary and secondary inductive. At 1 reactance and resistance at that particular generator frequency. At 1KHz, for example, this impedance will typically be many times the quoted impedance of the transformer so it is of limited use. What you really want to know is the inductance of each winding at 20Hz and 1KHz.

Cheers

Ian

 

[saint gillis]

So if we use this method we should switch the generator to 20Hz and do also the same measures to get something usable?

 

[ruffrecords]

So if we use this method we should switch the generator to 20Hz and do also the same measures to get something usable?

Yes and no. The problem is the inductive reactance is quite low at 20Hz so the error introduced by the dc resistance of the windings is greater but it will at least give you a rough indication. Also, the distortion at 20Hz will be a lot higher than at 1KHz so the impedance you measure will be altered by the harmonics produced. SO you need to make the 20Hz measurement at a much lower level; say at least 20dB lower.

Cheers

Ian

 

[user 37518]

A transformer is basically a pair of coupled inductors. The impedance of an inductor varies with frequency getting smaller as the frequency gets lower. This impedance forms a potential divider with the source. So, for a 150 ohm source like a mic, the impedance of the inductance should be no more than 150 ohms at 20Hz if we want the bass response to be no more than 3dB down at 20Hz. So this determines the number of turns necessary.

Also, the impedance of an inductor is directly proportional to the inductance, so a 10K transformer needs an inductance 10,000/150  = 67 times that needed for a 150 ohm one ( for the same bass response) which requires a lot more turns if the same core is used.

Cheers

Ian

How does the impedance of the inductor combine with the reflected impedances, say a 150:600 ohm transformer with a 10k ohm load in the secondary, that would transform the 10k load on the secondary, to a 625 ohm in the primary, is that 625 ohm in parallel with the 150ohm of the impedance of the transformer or what?

 

[ruffrecords]

How does the impedance of the inductor combine with the reflected impedances, say a 150:600 ohm transformer with a 10k ohm load in the secondary, that would transform the 10k load on the secondary, to a 625 ohm in the primary, is that 625 ohm in parallel with the 150ohm of the impedance of the transformer or what?

On the primary side, the source resistor forms a potential divider with the transformer primary inductive reactance. Over most of the audio bandwidth the inductive reactance is so high you can ignore it - all the source voltage appears across the primary. At low frequencies, the inductive reactance drops so the proportion of the source signal reaching the transformer drops due to the potential divider. Where the inductive reactance equals the source resistance the response falls by 3dB.

With a 150:600 transformer driven by an op amp for example, the source impedance is incredibly low so, even if the inductance of the primary is not very big, the low frequency response is good. For example, 1 Henry has a reactive impedance at 20Hz of 2 x pi x 20 x 1 = 125 ohms. SInce many op amps have an output impedance of a few tens of ohms, the drop at 20Hz even with a low inductance transformer is not very much. Some manufacturers take advantage of this to make smaller cheaper transformers.

It is a different story with tubes. The output impedance of a good tube output stage might be as low as 150 ohms but no lower and often the will not drive a 150 ohm load or even a 600 ohm one. My tube output stages will drive a 2400 ohm load and I use a 2:1 step down transformer to allow this stage to drive a 600 ohm load.For the transformer inductive reactance not to fall below 2400 ohms at 20Hz requires and inductance of 2400/(2 x pi x 20) = 19 henries

Going back to the original question a 150:600 transformer with a 10K load will look like 2500 ohms at the primary. This reflected resistance will appear in parallel with the primary inductance. The source now has to drive both the reflected impedance and the inductive reactance. Again over most of the frequency band the reactive inductance is much higher si it can be ignored and all we have is the low impedance of the transistor amp ( a few tens of ohms) driving 2500 ohms which is pretty easy and the loss is negligible. But still, at low frequencies, as the inductive reactance gets smaller than the 2500 ohms reflected load, the inductive reactance  begins to dominate the potential divider and so it still defines the low frequency response.

Bottom line is primary inductance at low frequencies is usually the primary determinant of low frequency response.

Cheers

Ian

 

[CJ]

sometimes they break it down into three equivalent circuits for low med and hi freq bands,

around page 476 is some good stuff>

https://babel.hathitrust.org/cgi/pt?id=uc1.b4154646;view=1up;seq=7

 

[ruffrecords]

sometimes they break it down into three equivalent circuits for low med and hi freq bands,

around page 476 is some good stuff>

https://babel.hathitrust.org/cgi/pt?id=uc1.b4154646;view=1up;seq=7

That's what you need. Fig 5b describes the HF behaviour, 5c describes the low frequency behaviour (which we were discussing ) and 5d is the mid band behaviour.

Cheers

Ian

 

[user 37518]

On the primary side, the source resistor forms a potential divider with the transformer primary inductive reactance. Over most of the audio bandwidth the inductive reactance is so high you can ignore it - all the source voltage appears across the primary. At low frequencies, the inductive reactance drops so the proportion of the source signal reaching the transformer drops due to the potential divider. Where the inductive reactance equals the source resistance the response falls by 3dB.

With a 150:600 transformer driven by an op amp for example, the source impedance is incredibly low so, even if the inductance of the primary is not very big, the low frequency response is good. For example, 1 Henry has a reactive impedance at 20Hz of 2 x pi x 20 x 1 = 125 ohms. SInce many op amps have an output impedance of a few tens of ohms, the drop at 20Hz even with a low inductance transformer is not very much. Some manufacturers take advantage of this to make smaller cheaper transformers.

It is a different story with tubes. The output impedance of a good tube output stage might be as low as 150 ohms but no lower and often the will not drive a 150 ohm load or even a 600 ohm one. My tube output stages will drive a 2400 ohm load and I use a 2:1 step down transformer to allow this stage to drive a 600 ohm load.For the transformer inductive reactance not to fall below 2400 ohms at 20Hz requires and inductance of 2400/(2 x pi x 20) = 19 henries

Going back to the original question a 150:600 transformer with a 10K load will look like 2500 ohms at the primary. This reflected resistance will appear in parallel with the primary inductance. The source now has to drive both the reflected impedance and the inductive reactance. Again over most of the frequency band the reactive inductance is much higher si it can be ignored and all we have is the low impedance of the transistor amp ( a few tens of ohms) driving 2500 ohms which is pretty easy and the loss is negligible. But still, at low frequencies, as the inductive reactance gets smaller than the 2500 ohms reflected load, the inductive reactance  begins to dominate the potential divider and so it still defines the low frequency response.

Bottom line is primary inductance at low frequencies is usually the primary determinant of low frequency response.

Cheers

Ian

Beautifully explained, I get it now, just one more question, the impedance rating of a transformer for example 150:600, is at which frequency?

 

[CJ]

oops, those equivalent circuits, refer to output transformers where we have a ton of current available to charge winding capacitors, thus, C is left out,

here are he equivalent circuits for input and innerstage, found around page 486 and later>

 

[user 37518]

Looks like a great book, thats just what I was hoping to find. Thanks CJ and Ian

 

[saint gillis]

Yes Ian great explanation of the effects of input inductance, I don't see transformers the same way anymore, the books goes a bit too far for me for now though..
So a 150:15K mic transformer, means that the transformer is optimized for 150ohm sources and that a 150ohm source will be reflected as 15k at the secondary ?

 

[CJ]

that's right, pick the primary to match the source, then pick he secondary to match the tube or transistor you wish to drive.

can you go too far with the primary inductance? yes, you will increase DCR with more turns, so you balance Henries versus signal chewing DCR. 
DCR can be brought down with a hi perm core that takes less turns to ge he same Henries, but then you have to watch saturation due to lower  flux level of hi perm core, a lot of things to balance, freq response, resonance,

 

[ruffrecords]

Yes Ian great explanation of the effects of input inductance, I don't see transformers the same way anymore, the books goes a bit too far for me for now though..
So a 150:15K mic transformer, means that the transformer is optimized for 150ohm sources and that a 150ohm source will be reflected as 15k at the secondary ?

That is right. It does not mean you cannot use other source impedances (lower ones will usually be OK) but you may not get the same performance.

For example, the 1:10 mic input transformers I use give a ruler flat frequency response with a 150 ohm source. However, they will often be used with condenser mics that can have an output impedance as low as 50 ohms. When you drive these transformers from a 50 ohm source you get a 5dB boost in the frequency response at about 25KHz.

Cheers

Ian

 

[saint gillis]

Ok, it's getting less foggy..

What about this 1:10 mic transformer (150:15k) : http://www.jensen-transformers.com/wp-content/uploads/2014/08/jt-115K-e1.pdf  -  the datasheet rates its input impedance 1.4K and its output impedance to 17K using a specific test circuit shown at the end of the doc, what are these values?