MXR Noise Gate issue- no output

[mpgoulet]

Hi, folks. This is my first time posting here. Long-time guitar teacher, first-time pedal builder. Please be nice!

So I have modeled one of the ToneLab circuits in Orcad to both learn the tool and understand the circuit. I cannot for the life of me seem to get any output from it at all, despite much banging of head against wall. I am posting a link to the schematic here in hopes that some kind person could take a look and see if anything obvious is wrong with it.



Also, the schematic this is based on is as follows:
http://www.tonepad.com/getFile.asp?id=35

Thanks!

[PRR]

> get any output from it at all

Same as debugging a real circuit.

This one is complicated, don't expect the whole shebang to work first try (maybe not 13th try).

Put your Voltmeter ( <V probe) at various places and see if the DC voltage is reasonable or way-wrong. Since this is a single-supply design with a Vb to bias things midway between zero and V+, expect most signal nodes on active stages to be pretty near Vb. Finding zero (or 0.01V DC) at an opamp in or out pin is way-wrong. The BJTs are less precise, but you expect B to be 0.6V higher than E, and C be somewhere between rails (gain stages) or C at V+ and E somewhere between rails (follower stages).

Everything DC-reasonable? Inject signal and use an AC probe to see the waveform at each point. V7 is feeding 300mV at reasonable frequency. Is that signal on the left end of C21? On the right end? How about at Q11 Base? And Emitter? From basic circuits, we would not be shocked if 300mV in gave 250mV at Q11 emitter: gain is "nearly 1" and there's some small extra drop in the input network. However 0.00mV would be wrong, and 0.03mV is very dubious (re-check the DC biases).

Just as in soldering, in e-schemetic layout it is often possible to have two leads at the same point yet not really connected. In my SPICE they sometimes overlap, look connected, but aren't. Dragging parts around sometimes reveals the slip.

One thing different in *some* SPICE compared to real parts: some implementations MUST see "Meg" for megohm. A bare "m" is ambiguous and some think "milliohm". Others are fussy "m" versus "M". In any cases, a million to one off the intended value usually screws the circuit.