transformer phase inverter ?

[billybillybilly]

ive got a couple old tube radio books from the 50's that have designs using a transformer as a phase inverter for a push pull output stage . it seems that they employed transformers for coupling between stages (RF-IF-AF etc) as well .
doing a little more looking , i found some old guitar amps also used this type of PI . i think one was a gibson .
anyway , any comments or suggestions on using a tranny based PI ?

[tubeswell]

More expense (for probably little extra benefit?)

[billybillybilly]

its also probably a tad bigger and heavier than a tube PI

but , it looked interesting in the sense that it isnt used alot (probably for good reason though).
i think the cons can be pretty easily identified (weight, size, actually finding one, etc), but does anyone know of any advantage of this type of PI ?

[stokes]

I have a 1:1 Triad transformer that I used in place of a PI tube and I found it to be kind of cold,sterile sounding and went back to the tube in that amp.I think it would be more beneficial in a HI-FI application.That is why you dont see it in guitar amps,as well as the space/expense consideration.

[Chuck H]

I thought about trying it and concluded that the cons out numbered the pros. JMHO though. An interstage transformer as a PI won't contribute any OD for canked tone, it introduces phase shifts that can complicate the feedback loop, it is more expensive than a tube circuit and it complicates layout to avoid hum.

The pros are: It won't distort (good/bad?), low impedance output and other than the limitations of inductance it won't contribute to THD.

The low impedance output was why I considered it. Using a pair of cathode followers after the PI in a real PITA. The transformer PI would give low impedance and relieve the circuit of two tubes and a handfull of circuitry. Looking at all this I decided working around the high impedance output of the typical LTP was better in the end.

Chuck

[Old Tele man]

...CBS-Fender used an transformer-PI in their super-duper 400W guitar amp.

[billybillybilly]

400w !?! holy moly , super duper is right !

i noticed in the schematics ive seen that there are no coupling caps between the PI and output tubes , like in the GA-20 . i wonder if the reissue has the tranny PI ?


thanks for all the input guys !
i appreciate your time !

[MWJB]

"...CBS-Fender used an transformer-PI in their super-duper 400W guitar amp."

And in the Musicmaster bass amp.

[Loogie]

I have a Gibson GA15RVT with one. I've often wondered why they were used. The part itself is more expensive than a tube, but you've also got consider socket, socket holes and heater wiring. With a transformer its a couple of holes and a few connections. Perhaps labor costs were part of the equation.

[hasserl]

"...CBS-Fender used an transformer-PI in their super-duper 400W guitar amp."

And in the Musicmaster bass amp.

The Musicmaster Bass amp sound great as a guitar amp. This is a good example of how a circuit using a transformer PI can work great for guitar.

[Enzo]

You can buy them from AES and elsewhere, not expensive. More than a 12AX7 certainly, but not expensive. As to why no one uses them, I'd also have to say money mostly. They weigh more than a tube, but so does a little reverb transformer.

I wouldn't worry about phase shifting, that's its job, and I doubt it will add any more unwanted shifting than a tube circuit. But I can also imagine it pretty difficult to tweak the curcuit. In a tube stage I can change caps and resistors easily - can;t easily change a transformer winding.

[loudthud]

The transformer PI makes class AB2 a lot simpler in the pre-semiconductor world. That means getting significantly more power out of any given set of output tubes. It also worked well for driving a pair of those big transmitting triodes that require drive in the 100's of volts.

[billybillybilly]

The transformer PI makes class AB2 a lot simpler in the pre-semiconductor world .

i guess that explains the lack of coupling condensers ... oops , i mean capacitors .

they seem to work similar to a reverb transformer huh ? aside from the secondary not being center tapped , primary gets the signal and B+ and secondary supplies in phase/out of phase signal to the springs ?

ive done quite a bit of tinkering with SS stuff (10+yrs) , new to the tube stuff , but obviously i have more research to do ... so basically a PI transformer would be a 300v 1:1 w/ center tapped secondary ?

[Enzo]

Well it works like a reverb transformer insofar as it IS a transformer, but the circuits are not alike. I offered reverb transformer as comparable in size and weight.

I don't know about 1:1. No reason to have that unless that is what you need. And the secondary doesn;t likely run at 300v.

Coupling caps are used to pass signal and block DC. If there is no DC in the secondary winding, there is no need for a cap. And likely as not, the bias supply would flow through the winding to get to the grids of the power tubes.

[Steve Conner]

The Musicmaster Bass and 300/400PS schematics are all available online. Just google them up to see how to use an interstage transformer.

In summary, you can hardly go wrong unless you include both interstage and output transformers in the NFB loop. Then it's time for some careful design. For instance in the 300/400PS, the OT was designed with a low-end bandwidth of 50Hz, but the interstage had to go down to something crazy like 5Hz to keep the NFB stable. I have one here and it's about the size and weight of a Deluxe PT.