why the reluctance to use an opamp buffer on a rat?

[m_charles]

Hello. Was doing some research, and I thought I would see this subject come up over and over again, but to my surprise, I barely found anything.
Was fiddling with a rat circuit, don't have any JFETs around (getting pretty hard to finnd those j201's these days.., but I have a bunch of op-amps. Seemed like an obvious thing to grab an unused half of an op-amp, put that to use on the output of the rat for the buffer, done.
Here is the example I'm doing (haven't breadboarded just yet). Obv this is coming off the volume pot.
Was mainly wondering if there is something I'm missing here? See any probs with this?
Thx!!!

[m_charles]

Sorry, should mention I wanted to isolate the output buffer from the circuit so added it after the volume pot instead of before like a stock rat. Will post the full schem when back at my big computer. shoulda done that in the first place.

[merlinb]

There's nothing wrong with that at all, apart from different tone stack loading

[antonis]

5532 (like most of bi-polar input op-amps) doesn't like high value bias resistors (like the 1M one)..

IMHO, 10k Volume pot could handle 100k bias resistor without issue..

P.S.
You can use the other half of op-amp to make a stiff +4.5V bias voltage divider..

[Fancy Lime]

Or connect the "bottom" of the volume pot to 4.5V instead if ground and leave out the bias resistor and the cap between the pot and the opamp.

[marcelomd]

5532 (like most of bi-polar input op-amps) doesn't like high value bias resistors (like the 1M one)..

Can you elaborate on that? I always use 1M and don't even think about it.

[antonis]

Can you elaborate on that? I always use 1M and don't even think about it.

It has to do with input bias current (and the respective voltage drop across bias resistor..)

For a 800nA max (1μA at full range working temperature) it might result into Vref 800mV "offset"..
(3.7V instead of 4.5V due to n-p-n inputs, hence current flowing into op-amp)

Of course, for FET input op-amps, you can use any high value bias resistor (if you don't mind about temperature instabilities due to elevated leakage current)..


P.S.
@Andy: Of course it can be done, at HPF's expense..
(although, I'm not fond of injecting AC garbage into DC point..)

[GGBB]

I don't think any "reluctance" would be primarily about using an opamp for the output buffer. It's about the LM308. Legend has it that the signature sound of the RAT comes from that specific opamp - which only comes in the single opamp flavour. Also - what's available commercially - which probably makes up the majority of DIY builds - is the standard topology. Trying to sell something a little different - especially if it doesn't support the LM308 - is not selling the RAT at all.

It should be fine with an opamp output buffer, but without the LM308 maybe it won't really sound like a RAT. You could always try two single opamps - say the LM308 and maybe a TL071. Or you could more easily use a BJT in place of the JFET.

[antonis]

Or you could more easily use a BJT in place of the JFET.

Couldn't agree more..

[Fancy Lime]

@Antonis
True. I normally dislike the whole "dumping AC on the bias point" as well. It may cause problems unless the impedances are chosen carefully and there is some isolation via an additional R and C. There may even be oscillation problems. On the other hand, it may not make a noticeable difference in a rat.
I for one would not miss the HPF in theis situation.

Andy

[antonis]

On the other hand, it may not make a noticeable difference in a rat.

As Gord said above, there are very few things (except its "heart") capable to alter its character..

[m_charles]

Can you elaborate on that? I always use 1M and don't even think about it.

It has to do with input bias current (and the respective voltage drop across bias resistor..)

For a 800nA max (1μA at full range working temperature) it might result into Vref 800mV "offset"..
(3.7V instead of 4.5V due to n-p-n inputs, hence current flowing into op-amp)

Of course, for FET input op-amps, you can use any high value bias resistor (if you don't mind about temperature instabilities due to elevated leakage current)..


P.S.
@Andy: Of course it can be done, at HPF's expense..
(although, I'm not fond of injecting AC garbage into DC point..)

Interesting, I had not thought of that at all. thx

[m_charles]

I don't think any "reluctance" would be primarily about using an opamp for the output buffer. It's about the LM308. Legend has it that the signature sound of the RAT comes from that specific opamp - which only comes in the single opamp flavour. Also - what's available commercially - which probably makes up the majority of DIY builds - is the standard topology. Trying to sell something a little different - especially if it doesn't support the LM308 - is not selling the RAT at all.

It should be fine with an opamp output buffer, but without the LM308 maybe it won't really sound like a RAT. You could always try two single opamps - say the LM308 and maybe a TL071. Or you could more easily use a BJT in place of the JFET.

Yes, I'm well aware of the op-amp mojo mysteries of the rat. "Slew rate" is everyone's favorite to talk about, haha. As far as the LM308, I have some around if I want to make a super accurate Rat clone. This is more just some experimenting for fun using the skeleton of this tried and true circuit. Main thing that brought up my post was that using the other half of a dual op-amp seemed so nice and tidy, that it seemed like something that many would have done before. Since I couldn't find much, I was thinking, damn, there's gotta be something I'm missing here.
But I would agree with you also that much of it probably is just a reluctance to change the architecture. I do know that the traditional JFET buffer can clip a bit, which some people find pleasing, so one of the trade-off is probably that the op-amp buffer sounds almost too transparent to some.

[m_charles]

Or you could more easily use a BJT in place of the JFET.

Couldn't agree more..

Not sure what's easier about that though. Less parts with the op-amp.

[m_charles]

Or connect the "bottom" of the volume pot to 4.5V instead if ground and leave out the bias resistor and the cap between the pot and the opamp.

This is actually a great idea. I'm always a fan of less parts! Going to try it.

[m_charles]

5532 (like most of bi-polar input op-amps) doesn't like high value bias resistors (like the 1M one)..

IMHO, 10k Volume pot could handle 100k bias resistor without issue..

P.S.
You can use the other half of op-amp to make a stiff +4.5V bias voltage divider..

the 10K pot was actually a mistake. I meant to put 100K pot there, haha

[m_charles]

5532 (like most of bi-polar input op-amps) doesn't like high value bias resistors (like the 1M one)..

IMHO, 10k Volume pot could handle 100k bias resistor without issue..

P.S.
You can use the other half of op-amp to make a stiff +4.5V bias voltage divider..

the 10K pot was actually a mistake. I meant to put 100K pot there, haha
Yes, I like using an extra op-amp for steady 4.5v, works great!

[antonis]

the 10K pot was actually a mistake. I meant to put 100K pot there, haha

It should be OK with 10k Volume pot, haha..

[antonis]

Yes, I like using an extra op-amp for steady 9V, works great!

Could you plz show us the particular op-amp configuration for stabilizing 9V..??

P.S.
Don't quote the above with something like: "The 9V was actually a mistake. I meant 4.5V there, haha"

[m_charles]

Yes, I like using an extra op-amp for steady 9V, works great!

Could you plz show us the particular op-amp configuration for stabilizing 9V..??

P.S.
Don't quote the above with something like: "The 9V was actually a mistake. I meant 4.5V there, haha"

I think I need more coffee, haha. yes, 4.5v