Driving an Octavia Transformer

[liquids]

I'm finalizing a transformer-based octave up design for myself, and I hit a datasheet/design brick wall.

Using the common 42TM022 txf, I'm driving the secondary, rectifying the primary, as ususal.

The data sheet says this txf's secondary's 'impedance' is 600 ohms, the 'resistance' is 45 ohm.... I'm driving said transformer with an op amp.  Holy crap, that's either very low, or really really low... So I realize, this is enough to torture your average op amp, no?  I mean, they design op amps for audio specifically to be able to drive 600ohm loads...which means your averate chip don't do it so hot.   

For one, should I be considering the load the op amp sees 600 ohms or 45 ohms?  It may be a moot point, but either way, that is torturous, no?  So how does the regular octavia circuit do it with just a dinky old BJT?  =D 

I also just looked at R.G's octave screamer, which suggests the TL002 transformer.  Said txf's secondary on that txf is 2k impedance, 200 ohm resistance.  A TL071 is driving it, so it must be functionally seeing a 2k load, no?  Or does no one else care about this kind of thing since we're talking nasty guitar pedals?

And okay, so maybe a TL07X op amp can handle 2k, and maybe 600 ohms unhappily, but 45 ohms?  A NE5332, etc, would be happier all around I suspect...with it's own tradeoffs....

What am I missing, or what else should I know so to speak?

Se, I concluded that I'd rather use a rail-to-rail op amp all around, but the datasheet says the LTC2262's output impedance is 220 ohms.  Boo. 

[R.G.]

The data sheet says this txf's secondary's 'impedance' is 600 ohms, the 'resistance' is 45 ohm.... I'm driving said transformer with an op amp.  Holy crap, that's either very low, or really really low... So I realize, this is enough to torture your average op amp, no?  I mean, they design op amps for audio specifically to be able to drive 600ohm loads...which means your averate chip don't do it so hot.   

You're missing an important fact that most people miss when they first look at transformers.

Transformers don't have impedances, they have ratios.

What 1.5K to 600 ohms really means for this transformer is "if you put a 600 ohm resistor on the secondary winding, whatever is driving the primary will think it's driving a 1500 ohm resistor, at least within the limited frequency range of 300Hz to 3400Hz." What happens outside the pass band is not guaranteed, but we know it will go to 45 ohms for the wire resistance at DC, and will vary wildly as you get way above the upper range.

For one, should I be considering the load the op amp sees 600 ohms or 45 ohms?  It may be a moot point, but either way, that is torturous, no?  So how does the regular octavia circuit do it with just a dinky old BJT?

Neither. The opamp only sees 600 ohms if the primary-used-as-secondary has a 1500 ohm resistor on it. What it actually has is some diodes which dump current into a resistance alternately. When the diodes are on (they have to get to more than their turn on voltage) the primary sees 600/1500 times the resistance the diodes are driving. I don't have the schemo in front of me, but that's pretty high as I remember, much more than 1500 ohms, so the load on the 600 ohm side that the opamp sees rises accordingly.

There are, as you note, very few opamps which can drive 600 ohms well and with any grace. Yes, it is some torture to the others if the load is really 600 ohms. And the dinky old BJT is probably capable of driving much more power than the opamp; it can certainly do much more current. The problem with a discrete BJT is it's not nearly as linear or predictable as an opamp.

[liquids]

Thank You.  I will sleep better now. 

[PRR]

When you play ZERO Hz, the opamp feels 45 ohms.

The specs probably suggest the impedance rises above 600 ohms by 300 Hz. Yeah, below 200 ohms at the bottom of the guitar band. If it swings 1V peak in 200 ohms the opamp must deliver 5mA. Any general-purpose opamp will do that. Maybe not good enough for precision government work, but what are you doing NEXT? You slice the fish down the middle, flip it and stack it for a double-decker. Whatever "damage" the opamp may do around the edges is nothing compared to that.

> I'd rather use a rail-to-rail op amp

Why? "Rail" is not a good place for audio. It's all the same issues, except (with light load) it could swing 8.99V instead of 7V.

[liquids]

Why? "Rail" is not a good place for audio. It's all the same issues, except (with light load) it could swing 8.99V instead of 7V.

Exactly.  Maximum clean gain available into the transformer.  Make that swing 9.39v thanks to Visual Sound.   And I never find much use for those r-t-r op amps I have laying around. 

I was going to do +/-8v charge pump but I'll settle for 9.39v clean swing if it saves me space.  It's a guitar after all.