stacking op amps to increase current

[mordechai]

If I am using clipping diodes that require a higher current before they clip (like LEDs), should stacking two 4558 chipcs compensate and provide them with enough to clip more closely to the way a silicon diode clips with a single 4558?

[Mark Hammer]

I'm still not sure if stacking, in the sense of soldering the pins together and plugging the piggyback result into a socket, is the optimal approach.  I know it sounds different, because I've experimented.  But what a person may really want to do is have two separate op-amps, wired up in a manner similar to what you see used for NE5532-based headphone amps or Jack Ormans Super BUffer.

So, something ELSE produces the gain, and the signal then passes through several independent-but-ganged unity-gain op-amp stages, whose output current is summed.

Alternatively, if there is a series resistor just ahead of the clipping diodes, one can simply vary the resistance to set the amont of current hitting the diodes.

[gritz]

How do you mean "higher current before they clip"? Clipping is a voltage thing - it starts when the applied voltage is greater than the forward voltage of the semiconductor junction and allows current to flow - say about .6 volt for a silicon diode and about 1.7ish volts for a red led. The amount of current through the clipping semiconductors is defined by the topology of the circuit (input / feedback resistors, etc), but pretty much every design you'll see won't trouble any typical opamp. Do you have a pic of / link to the circuit you're using?

[mordechai]

Mark and Gritz, thanks for the replies.  Yes, I meant voltage, not current -- apologies.  I am about to begin tinkering with a BYOC TS-808 kit so was thinking about ways to get LEDs to clip just a BIT more like silicons.  Mark, your comment about the series resistor is interesting -- in the TS-808, that would be the 4.7K to ground, right?  Whereas on something like the DOD OD250, it would be the 10K in front of the diodes, correct? 

[Mark Hammer]

I honestly don't know what would modify the current to the diodes in the TS-808 topology.  However, in the case of the Dist+/DOD250/Rat, it IS the series resistor before the diodes.

[mordechai]

Well -- I've toyed with the OD 250 values (diode types, resistor before the diodes, cap before the resistor) and here's what I found:

Using red asymmetrical LED clippers offers a very nice, less "hard" type of clipping, with less distortion -- which I actually like a lot, and the output wasn't nearly as compromised as when I tried using germaniums (as per the "Liquid Drive" on Fuzz Central).  Mark's "proximity to clip" concept seemed to be in full force here, and it worked very well.  I should note that I did not try this with regular symmetrical LEDs (one "up", one "down"), so I'm not sure how that would affect tone or volume.

Reducing the resistor from 10K to 8.2K allows for more distortion to be coaxed out of the LEDs, though the noise level increases a bit.  Reducing the 10uF cap in front of the resistor to 4.7uF tightened it up a bit more too without cutting out too much. 

I would definitely like to incorporate these options into a TS type circuit.  In that scenario, I would think that the 1M in the feedback loop could be toyed with for changing the response of the clipping diodes -- any thoughts on this?  I'd like to get some opinions before I buy the parts to put together a TS circuit...

[mordechai]

Just applied some of the aforementioned changes to a TS-808 circuit.  I used red LEDs in place of the usual silicon diodes, and lowered the resistor to ground from 4.7K to 3.3K.  This gained some of the grit/dirt back into the clipping, but the resulting overdrive sounded less compressed, more natural, and there was a nice increase in available output.  I A/B'd this against my original TS-808 with stock values, and I think I prefer this modification...