Jensen AN-002 App Note Transformer Question

promixe · Jul 27, 2011

Hi All,

I've been reading Jensen AN-002 to understand how impedances reflect between windings (page 2). My question is about the input transformer impedance section:

we need to know what the transformer secondary "looks like", impedance wise, to the amplifier. If we assume that the primary is driven from the line in our previous output transformer example with its 80 Ohm source impedance, we can calculate that the secondary will "look like" about 225 Ohm to the amplifier input.

How exactly does he arrive at 225 Ohm figure?

The part above it:

We can calculate that this load will "look like" about 42 kOhm at the primary

..I think I understand: 2k43 * 16 + 2k5 + 92 = 41k5 (~42k)

But the 225 Ohm I can't reverse-calculate. Thanks for your help!

Jean Clochet · Jul 27, 2011

promixe said:
How exactly does he arrive at 225 Ohm figure?

The part above it:

We can calculate that this load will "look like" about 42 kOhm at the primary

Click to expand...

..I think I understand: 2k43 * 16 + 2k5 + 92 = 41k5 (~42k)

But the 225 Ohm I can't reverse-calculate. =

What do you get?

First, 2K43 +92 X 16 + 2K5
And, 80 + 2K5 / 16 + 92 in // with the 2K43 load R.

promixe · Jul 27, 2011

Ok, but that gets me 253R, no? I think you meant (80+2k5) / 16 + 92, right?

I actually did calculate it like that before but since it's not quite 225R I got confused and thought this wasn't the right way.

Jean Clochet · Jul 27, 2011

When you got your 253R answer, did you then add in the parallel of the 2K43 load resistor on top.?

promixe · Jul 27, 2011

Jean Clochet said:
When you got your 253R answer, did you then add in the parallel of the 2K43 load resistor on top.?

Aha! =) Gotcha, that gets me 229R then. Sorry, didn't pay attention to your "//" sign there. =)

So, just to verify the math here, if I'm calculating this for Carnhill VTB9045 (1:2 mic transformer) with both primary and secondary windings in series:

Each primary winding DC resistance is 24R and each secondary is 130R. The mic source is 150R and the secondary loading resistor is, say, 5K. Therefore: ((150R + 48R) * 4) + 260R = 1052R || 5K = 870R. This is what the amplifier stage will see as its source when a mic plugged in through this transformer. Correct?

JohnRoberts · Jul 27, 2011

promixe said:
Hi All,

I've been reading Jensen AN-002 to understand how impedances reflect between windings (page 2). My question is about the input transformer impedance section:

we need to know what the transformer secondary "looks like", impedance wise, to the amplifier. If we assume that the primary is driven from the line in our previous output transformer example with its 80 Ohm source impedance, we can calculate that the secondary will "look like" about 225 Ohm to the amplifier input.

Click to expand...

How exactly does he arrive at 225 Ohm figure?

The part above it:

We can calculate that this load will "look like" about 42 kOhm at the primary

Click to expand...

..I think I understand: 2k43 * 16 + 2k5 + 92 = 41k5 (~42k)

But the 225 Ohm I can't reverse-calculate. Thanks for your help!

Apparently it isn't exactly 225 ohm (as the problem is defined). Which makes it a less than optimal example when trying to inform.

first, you need to add the 92 ohms to the 2430 before the x16 +2500 so= 42,852 (closer to 43k than 42k)
looking the other way ( 80+2500)/16 + 92 || 2430= 229 ohms again not very close to 225

While a sloppy example for teaching purposes, I suspect the error may not be in his calculation or result, but perhaps in the rounding off one or more of the several numbers used in that example.

I'm not sure who needs 4:1 step down input transformer, but now we know the math.

JR

promixe · Jul 27, 2011

Thanks for clarification JR.

Another question. Here is a simplified circuit of a mic going into a low noise BJT preamp (THAT1510/1512/1570 and alike) through a mic input tx:

With the extra resistors (R1+R2 for phantom and R9+R10 for input bias) how do I calculate the same things?

From my post above about calculating the VTB9045 impedances:

- the mic will see (5k + 260R) / 4 + 48R = 1363R. I "think" that the pair of 6k81 resistors would be in parallel with primary DC resistance, correct? Then it would be negligible difference in calculation (~0.2R). Do R9+R10 need to be in parallel with 5k?

- the preamp will see ((150R + 48R) * 4) + 260R = 1052R || 5K = 870R. Then in parallel with R9+R10 it would be 639R. Where do R1+R2 come in play in this calculation?

Sorry for the confusion, just trying to finally get a clear picture of this topic once and for all. =)

Jean Clochet · Jul 27, 2011

promixe said:
Each primary winding DC resistance is 24R and each secondary is 130R. The mic source is 150R and the secondary loading resistor is, say, 5K. Therefore: ((150R + 48R) * 4) + 260R = 1052R || 5K = 870R. This is what the amplifier stage will see as its source when a mic plugged in through this transformer. Correct?

Yes this is what the amplifier input will see, you've got it

Sorry about not being clearer about the // symbol.

Yes, as John said about the Jensen being a bit sloppy in the sums department. Although, within striking distance is good enough for practicality.

Jean Clochet · Jul 27, 2011

Didn't see this post before I reply

promixe said:
- the preamp will see ((150R + 48R) * 4) + 260R = 1052R || 5K = 870R. Then in parallel with R9+R10 it would be 639R. Where do R1+R2 come in play in this calculation?

R1 and R2 would be in // with the 150 ohm source before you add in the DCR and square the turns up etc. So, as you said with the other way around, it's negligible. I'd say you have a clear picture on it

Edits: if you don't need the extra 6dB or need a lower source for the amp for noise reasons, just use the secondary in // as a 1:1 with 1K2 load resistor.

promixe · Jul 27, 2011

Jean, thanks for the clarifications, it's making more and more sense now. =)

What about R9+R10 in:

promixe said:
- the mic will see (5k + 260R) / 4 + 48R = 1363R ...Do R9+R10 need to be in parallel with 5k?

Thanks!

Jean Clochet · Jul 27, 2011

promixe said:
What about R9+R10 in:

promixe said:

- the mic will see (5k + 260R) / 4 + 48R = 1363R ...Do R9+R10 need to be in parallel with 5k?

Click to expand...

Thanks!

I don't know the amp you are using. It looks like they are just providing a 2K4 load for mics though. Why not just split the 5K in half instead and use that? Otherwise, yes they need to be in // with the 5K for calculations like this. But, in your case, then you are putting a much lower load on the traffo secondary and , by turns, the mic so, take them out and split the 5K. Or leave them in but go up considerably in value so they are negligible.
Edits: Or ground the secondary centre-tap, leave the 5k as is and remove the two 1K2's. Lots of ways to do the same thing.

promixe · Jul 27, 2011

I see. R9 & R10 can't be removed because they are biasing the input BJTs in THAT15xx preamps. But...

On page 5 of THAT1570 Datasheet in "Improving Noise Performance" it says if the bias resistors are to be increased they may significantly increase noise.

Also Jensen states that:

To determine the proper ratio transformer for your application, you will need to know the Voltage Noise and Current Noise of the active amplifier device to which the secondary of the transformer will be connected. If you divide the Voltage Noise by the Current Noise of the device you will get a value in Ohms. This value is the Optimum Source Impedance for best noise performance using that particular amplifier device. Select the transformer whose secondary impedance most closely matches the calculated value.

According to the above my optimum source impedance would be 500R... Hmm.. =)

JohnRoberts · Jul 27, 2011

promixe said:
Thanks for clarification JR.

Another question. Here is a simplified circuit of a mic going into a low noise BJT preamp (THAT1510/1512/1570 and alike) through a mic input tx:

With the extra resistors (R1+R2 for phantom and R9+R10 for input bias) how do I calculate the same things?

From my post above about calculating the VTB9045 impedances:

- the mic will see (5k + 260R) / 4 + 48R = 1363R. I "think" that the pair of 6k81 resistors would be in parallel with primary DC resistance, correct? Then it would be negligible difference in calculation (~0.2R). Do R9+R10 need to be in parallel with 5k?

- the preamp will see ((150R + 48R) * 4) + 260R = 1052R || 5K = 870R. Then in parallel with R9+R10 it would be 639R. Where do R1+R2 come in play in this calculation?

Sorry for the confusion, just trying to finally get a clear picture of this topic once and for all. =)

I will ignore the obvious question (why?)..

From the carnhill website series primaries are 1200 ohm, series secondaries are 4800 and turns ratio are 1:2

so mic will see (5k || 2k) + 4800) /4 || 13.6k so approx 1.4k a fair termination for a mic.
the IC will see ( ((150 ohm || 13.6K) + 1200 ) x 4 ) +4800 ||5k||2K.. approx 1.2k almost 10x a typical mic

So you have 2x transformer step up, but 1.2k effective source impedance feeding a IC optimized for 150-200 ohm source. If you are in love with the transformer there may be a more optimal IC, if you are in love with the 1510, why use a transformer? Note: I haven't done the math on the 1510, just speculating based on how it is designed to be used.

JR

Jean Clochet · Jul 27, 2011

I suspect he wants the sound of the Neve grafted onto the chip.
Alternative: Strap traffo as 1:1, leave the 1K2's as they are and replace 5K shunt with a 2K4 shunt. Source z will be lower than your target ~500 but probably OK too. The chip was expecting a direct mic source anyway wasn't it? So this will about that.

Jean Clochet · Jul 27, 2011

JohnRoberts said:
the IC will see ( ((150 ohm || 13.6K) + 1200 ) x 4 ) +4800 ||5k||2K.. approx 1.2k almost 10x a typical mic

Not quite. We don't want to add the reflected primary 1K2 before multiplying by four. We only add the DCR which is 48. And the same with the secondary 4800. That bit should be the secondary DCR which is 260.
Anyway, we end up with about ~600R

promixe · Jul 27, 2011

Jean Clochet said:
I suspect he wants the sound of the Neve grafted onto the chip.

Yes, I'm investigating the sound options with various transformers going into a clean preamp and/or remote controlled preamp (THAT1570).

Alternative: Strap traffo as 1:1

How do I do that with VTB9045?

Jean Clochet · Jul 27, 2011

Sounds good to me

promixe said:
Alternative: Strap traffo as 1:1

Click to expand...

How do I do that with VTB9045?

Strap 7 to 8 and also strap 9 to 10. Signal comes from 7 &8 (+) and 9 & 10 (-).
There is usually a little pF cap on the Neve traffo for RF filter. It's in // with the shunt.
Usually it's a 180pF if I'm right - it's been a long long time since I looked.
You will want to multiply that value by 4 for the same filtering at a 1:1 strap.

JohnRoberts · Jul 27, 2011

My bad I didn't see the DCR listed below

I am not the best to ask about transformers as efx .

Good luck.

JR

Jean Clochet · Jul 27, 2011

JohnRoberts said:
My bad I didn't see the DCR listed below

I am not the best to ask about transformers as efx .

Good luck.

JR

Well, I think Mr Bill Whitlock of Jensen is supposed to be the one to ask about transformers but he got it wrong too: 225R vs 229R

I do stuff in my head all the time and mess up frequently by a decade. It happens.

Yes, good luck Promixe.

promixe · Jul 27, 2011

Oh I see, basically primaries in series and secondaries in parallel. The the mic will see 1k3 and the chip will only see 216R.

Thank you guys!

So to summarize: for THIS PARTICULAR application (i.e. into THAT15xx) it's best to use 1:1 transformers for proper mic loading.

I wonder how Owel (FiveFish) does it with 1:8 Cinemag in his SC1-mk3 THAT1510-based pre...

Jensen AN-002 App Note Transformer Question

Help Support GroupDIY Audio Forum:

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Well-known member

Similar threads

Join the conversation!

Join today and get all the highlights of this community direct to your inbox. It's FREE!