Here's a skeezer-teaser for you EE's out there...

[jbeckva]

Ok, so I had to put my PIP card (aka basically a unbalanced to balanced converter) back together tonight because totally bypassing it "weren't such a good idea". Isolation transformer = gooooood, yah! So, back to the drawing board a lil' bit.

Anyway, first look at the left side on the input. That was a barrier strip, but now I have it as an RCA jack with the signal and chassis grounds tied together. Now look at the far right, and see the opamp set up as unity gain with the attenuator pot feeding the signal in.

In between the output of the isolation transformer and that final buffer is two things. First, a high pass selectable filter that could be switched in to filter low freqs at 3 cutoffs. It also has some sort of defeatable "compression horn" EQ that would put a boost on the high end. So from there... the output of the filter/eq stage goes into a 23khz low pass filter stage, then finally the buffer output.

I would like to get rid of both filter circuits, and just hook the secondary straight to the buffer amp, or get rid of it altogether. I don't like the idea of unnecessary "stuff" in the signal path between the preamp and amp. Anyone have some good pointers or suggestions?

[attachment=0:228s22gi]pipschem.jpg[/attachment:228s22gi]

 

[laatsch55]

What's the impedance of the input op-amp buffer got to be??

 

[jbeckva]

Not sure. Start with the 1st op amp stage (U2-A) from the left. JP2 (the horn EQ switch) is open, so R21 and C11 are not in-circuit. The transformer is a 1:1 isolation transformer. It has 20K across it's output/secondary. R23 is the feedback transistor at 5.36K but it is also tied to ground with another 24.9K of resistance via R22. What's the gain of that stage going into the 3 position hi pass filter circuit?

From there, U2-B is set up as unity gain pretty much, but there's also that network of C16, 17, R34 and 35 which is the low pass 23Khz filter.

What I was thinking of was taking all of U2 out of the circuit completely, and simply hooking the high side of the transformer's output straight into R48's high side, leaving R48's attenuator wiper still in to control the overall output via U3's unity gain buffer. But to do so, do I have to (for example) change R20's value to match the input of U3 properly? All goes back to the 1st amp stage... what's the gain of that?

 

[jbeckva]

Ahhh... Ok, so the manual for the PIP says it has a 20K input impedance, so maybe it's safe to assume that since it's a 1:1 it presents a 20K load on the output. But with R20 also at 20K that is cut to 10K. 10K is also what happens to be the value of the attenuator across the buffer's input, so there would be a decent match there, correct? (or am I nuts?)

 

[laatsch55]

I think ya got it Jer.,

 

[Gepetto]

Hi Jerry
Remove R20, 20K
Take U2 out
Leave R36 jumper in the open position
Change R40(?? hard to read) from 10K pot to 20K pot
Jumper T2 transformer pin 5 to the top of R40(?? again hard to read)

That should give you a transformer isolated signal path with unity gain (or close to it) with attenuation if you desire.

If you want to get fancier and provide boost as well, we can figure that out too.

 

[jbeckva]

How necessary is it to change out the attenuator from 10K to 20K?

 

[Gepetto]

If you don't change it Jerry you will see 10K input impedance instead of the 20K that was there prior. If you do not care about the 10K input impedance, leaving the 10K is fine.

 

[jbeckva]

If you don't change it Jerry you will see 10K input impedance instead of the 20K that was there prior. If you do not care about the 10K input impedance, leaving the 10K is fine.

Thanks so much - that worked out PERFECTLY. Now all I have in the path is the transformer. For the buffer.. drum roll... I have a LM4562, which kicks butt for S/N and distortion. I could tell the difference immediately.

 

[avionic]

Need to hang a couple big ass capacitors on the back of it.