Simple + Cheap Solution for Driving a Transformer?

Hi,

I am thinking of using this circuit from Jensen:
http://www.jensen-transformers.com/as/as025.pdf

Not thrilled about the $8 opamp though.

This is the transformer I'm using:
http://www.jensen-transformers.com/datashts/123slcf.pdf

I've looked around in this forum for ideas but mostly found things too complicated or expensive for the time constraints and budget of my project. I need something I can whip together fast and hopefully with parts I have on hand.

On hand goodies at the moment are as follows.
LME49860
OPA604
OP275
BD139 & 140
NTE128P & NTE129P
BC556A
BC547B
Some household 2NXXXX devices.

Any suggestions? Pointers? Tips? Things to be wary of?
I really appreciate any help.

opamp driving a bd139/140 totem pole to drive the tranny?

EDIT: like figure 11 here:
http://www.headwize.com/projects/opamp_prj.htm

Thanks Svart!
figure 11c looks almost ideal. Since I'll power it from +/-18V I imagine I would want to change the resistor values. The author doesn't mention how he calculated those resistor values. Obviously, I have no battery related constraints so I'm not concerned with low current draw.

How do I calculate the values here for my application?

Looking at the specs for the transformer it seems like I'll want to keep the source impedance as low as possible. See page two upper right side chart.

How low could I go using this topology?

[quote author="Svart"]opamp driving a bd139/140 totem pole to drive the tranny?

EDIT: like figure 11 here:
http://www.headwize.com/projects/opamp_prj.htm[/quote]
Or for an off-centre position for the opamp, 'enter' the diode-stack on top, not in the middle. Should give less cross-over of the opamp. While it can be measurable, it might not really be noticable, but let it be said.

[quote author="clintrubber"]
Or for an off-centre position for the opamp, 'enter' the diode-stack on top, not in the middle. Should give less cross-over of the opamp. While it can be measurable, it might not really be noticable, but let it be said.[/quote]

Do you mean like this?
http://www.jlmaudio.com/JLM Hybrid Opamp circuit.pdf

This circuit would be very convenient since I have the necessary parts on hand. Although, it does mention a minimum load of 50 ohms. I see 20 ohms in the datasheet for the transformer so with the primaries in series I'm still 10 ohms below the Hybrid spec. would it be bad to put 10 ohms in between the driver and the transformer?

I apologize if this is a silly question...

> Not thrilled about the $8 opamp though.

It's a $55 transformer!!! Why strain about another eight bucks?

If you just can't afford $8.... the transformer does not need that fancy driver. Skip the NFB from transformer and just smack it with a $1 NE5532 in any ordinary low-gain connection.

The OPA604 and OP275 are no dogs neither. Unless you have honest 600 ohm loads out there, either will do fine.

In the old days we just hung 47 ohms off a uA709 IC into a $10 transformer for all but our Main outs.

> mention a minimum load of 50 ohms. I see 20 ohms in the datasheet for the transformer

20 ohms at DC! How much DC do you listen to?

That transformer alone is over 600 ohms by 20Hz. If loaded with honest 600 ohms it will be 600 over the audio band; if not, it will be higher over most of the audio band. Stray DC is a minor problem for both opamp and transformer, but JLM's plan has a servo which will make DC error about zero.

[quote author="PRR"]>
It's a $55 transformer!!! Why strain about another eight bucks?

If you just can't afford $8.... the transformer does not need that fancy driver. Skip the NFB from transformer and just smack it with a $1 NE5532 in any ordinary low-gain connection.

[/quote]

After four channels the opamp adds up to more than the budget for the project allows. This is not my own project, I am racking up some channel strips for a friend and unfortunately he's not a rich friend :?

I like the idea of the regular low-gain driver. It is nice and simple. I borrowed this from a Jung application note.... (redrawn)




I'm thinking I'll use either the OP275 or the LME49860 I have leftover from another project. Both have fairly low offset.

In the old days we just hung 47 ohms off a uA709 IC into a $10 transformer for all but our Main outs.

What did you do for your main outs 'in the old days?'

[quote author="somnium7"]Hi,

I am thinking of using this circuit from Jensen:
http://www.jensen-transformers.com/as/as025.pdf

Not thrilled about the $8 opamp though.

[/quote]

I agree with you, you may choose some more expensive opamp, or even class A discrete amp, if you don't want to spoil an expensive transformer.

The parts for a FETbloke are less than 8$ and will drive that tranny..

[quote author="Svart"]The parts for a FETbloke are less than 8$ and will drive that tranny..[/quote]

I read through all the threads that came up in search for FETBloke. Looks very cool. However I can't figure out which is the current version...

Anyway, thank you all for all the info. It is quite a bit to digest. I do feel like I have a better idea of where to start.

they were all current versions, just different flavors of the same thing. I built mine with the JFET front end, current diodes and darlington outputs. They are great.

[quote author="mediatechnology"]Roger and I have had pretty good luck using a THAT1646:
For the maximum slew rate the 1646 can provide, 15V/uS, I'd stick with the differential drive connection shown in the drawing above. The paralleled common mode drive circuits are ~1/2 that slew rate. This driver solution is <5$ and performs very well.[/quote]

Interesting! I briefly looked into using this IC but the person I bought the transformers from said it was a bad idea.
I have placed an order for some Bloke boards for this application. Once I have them built up I'll power up my signal generator and make some comparisons between them and the Jung circuit above and decide which works better.

However, I might consider using the 1646 in my console which I am overhauling. It already has nice DOA based output drivers for most outputs but I really want to convert the unbalanced send/receive loops to balanced and space is very limited....

Thanks!!!

I've looked/listened at the 1646 driving a Jensen 600:600, and driving an output directly, and it was able to handle the transformer fine with no apparent sonic changes. This may sound counter-intuitive, but I left the Jensen out of the circuit in question because it added no significant sonic fingerprint. To me this says the 1646 has no problems driving the transformer, and the Jensen in question was very clean.

[quote author="mediatechnology"]Roger and I have had pretty good luck using a THAT1646:



[/quote]

:thumb:
+1 from me. I've been using something similar based on the THAT data sheet. It's a very nice chip to have around.

really interesting thread.

and may I ask, where can I buy That chips in Europe?

thanks

Cool,
I was looking for them for a while.

thanks