HEP-722 available equivalent

[CeaSaR]

... mistake from the OP ...

Yep, I made a mistake. Not sure where I got the PNP from. I thought I read it in the body
of the article, but it's not there. If you you look at the time stamp, you'll see it was on the
late side (or early as it may be called). Ah, well. C'est la vie.

NPN, yes it is. Still not enough time to play with the sim yet. Perhaps a little later.

As for the date of the article, the web owner doesn't say, but I'd put the lingo around the
mid '70s.

CeaSaR

[MrAl]

Hi again,


Yes it is an NPN and i gave a couple of cross references to a part with similar spec's that my old old cross reference list gave.

I checked out the twin tee filter on the output, and yes it's a notch filter and the pot varies the placement of the notch with frequency and also the phase shift. I was guessing that with the right adjustment the phase shift would be enough to cause oscillations (as in the writeup) but it also needs to have enough gain at that point too, and this might be too low. As the pot is adjusted lower, the phase shifts more, but the gain goes down too, but it's more complicated than that so i guess i'll have to look at it some more.
I did get it to 'ring' when i changed the filter to a more true twin tee but the oscillations damped out quickly anyway.

[sofaspud]

My NTE book shows a direct cross from HEP722 to NTE108. A quick Google search brought me to Newark's web page for a
Fairchild 2N5770, with the following specs:
# RF TRANSISTOR, NPN, 15V, TO-92
# Transistor Polarity:NPN
# Collector Emitter Voltage, V(br)ceo:15V
# Power Dissipation, Pd:350mW
# DC Collector Current:50mA
# DC Current Gain Max (hfe):200
These pretty much agree with the NTE108 cross-ref listing.
Perhaps the 2N2222s, 2N3904s, NTE123s won't work well in this circuit. Hence the use of the HEP (Hobby Experiment Prototype,
something like that) transistor. 2N3904s were common project parts back in the mid-70s.
And of course, this is a stomp box circuit. Particulars aren't as relevant as the tone.

[MrAl]

Hi,

What do you think is so different about the 2N3904 that it shouldnt work in this circuit?

[dyarker]

hfe half, or less, of HEP-722.
Article says to adjust just to point of oscillation, then back off a bit. 2N2904 may not have enough gain to oscillate.

Two 2N3904s in Darlington would be WAY TOO MUCH gain.

Two 2N3904 amps cascaded could be built with right gain; but parts count up, and phase would be wrong.

Three 2N3904s, getting rediculous. Might be better off Googleing for an op amp circuit that does same thing.

On the bright side, we don't know a 2N3904 won't work ... yet

[CeaSaR]

...
Try PN2484.
or NTE108
NTE123AP
...

MCM has the 108 and 123 in stock. No 2484 found. Check out the search.

Now to get my order together.

CeaSaR

[sofaspud]

Yes, from my look-see the 2N3904 and her relatives didn't appear to have enough gain. And as dyarker noted also, a
Darlington configuration is probably overboard, given with the low-power signals the circuit operates at. MrAl mentioned a
2N3866 and that might do the job, though it might not. That's why I added that last bit; I don't think any specific
parts are important, just that the circuit has enough gain to twang or funk or whatever.

[MrAl]

Hi,


I tried first a 2N3904 and it didnt work, so lucky in Spice we can increase the gain anytime we want, so i increased the gain by 2x and that didnt help. Increased again another 2x (now it's 4x original) and that didnt help, increased 2x one more time (now that's 8x original gain) and that didnt help either.

[dyarker]

Odd
C6 is connected to base, and you varied R4 from 2K down to zero?

It should oscillate. Hmmm, on simulator input would PERFECTLY quiet. Can you put one pulse on input to give it kick start?

Cheers,

[MrAl]

Hi again,


Yes this is odd, as i expected it to oscillate somehow, some way, yet i tried everything. Yeah, i tried using a step input and also sine source that would look like a guitar. I also tried doing it with large and small input resistance, smaller input cap, even a near 'open' input such as you would get when the guitar is disconnected. I thought the writeup in that article was correct but maybe not, or else something just isnt right.
I also tried varying the resistor values (every one) and cap values (every one) and did an impedance scaling on the twin tee network just in case it did not provide a low enough impedance to the input to oscillate. Still nothing.

The only thing that did have an effect was changing the values in the twin tee network so that they are more in line with a standard notch filter. That produced a ringing, but it damped out.

I analyzed the twin tee section separately just to see how it responds, and found that it is a notch filter, and can shift the phase quite a bit (enough i would think for oscillation) but as the phase shifts more the gain drops, which is contrary to how you would want to make an oscillator. That puzzles me too.

I was thinking of doing a full blown analysis of the circuit as a feedback system, and see what values have to be changed to get this to oscillate if it ever does but that involves some time to complete. I thought that by fooling around with some of the values it would eventually oscillate in spice, but that doesnt seem to be possible or else the combination that it takes is just too strict. From the writeup though you wouldnt think so. Change R4 from 0 to 2000 and somewhere it should oscillate, but it doesnt. I wanted to use that as a starting point to see the waveform when a 'guitar' string is plucked, but i cant get it to that point yet and supposedly that's the way it has to be done.

Any more ideas i'd like to hear too.

[dyarker]

I don't play git-tar; but I think damped ringing is the objective; that's why the text says back off when oscillation starts.

Cheers,

[MrAl]

Hi again,


Yes, but that's not the point i was making. The writeup says it does oscillate and that's an indication that you are close to the optimal adjustment. That's how i would know how to adjust it in real life or in simulation. If it doesnt oscillate, i dont really know what the best adjustment is. Yes it might still work, but i cant be sure it is working the way the author intended because it wont oscillate like he said it would.

[Bob Scott]

The writeup says it does oscillate and that's an indication that you are close to the optimal adjustment.

Well, there could be a mistake in the schematic. It is common for a magazine art department to makes mistakes. THEY wouldn't recognize an error. They were arts students.

Maybe there was stray capacitance in the prototype?

[Joseph]

Has anyone built the circuit in LTspice for playing around with?

[MrAl]

The writeup says it does oscillate and that's an indication that you are close to the optimal adjustment.

Well, there could be a mistake in the schematic. It is common for a magazine art department to makes mistakes. THEY wouldn't recognize an error. They were arts students.

Maybe there was stray capacitance in the prototype?

Hi Bob and Joseph,


Bob:
Well i assumed that the stray capacitance would not affect that circuit that much in the audio range, where this should oscillate.

Joseph:
I can draw one up if you would like to play around with it for a while. I've been using another simulator.