cabinet simulators for direct recording.

[armstrom]

I've been looking for a decent (yet simple) speaker/cabinet simulator circuit to integrate into a little preamp project I'm working on. I want to keep the part count low and the board small and am willing to compromise some (ok, quite a bit!) of accuracy to do it. In other words, I want something passable, not perfect. I've come across a couple of designs that seem to fit the bill but have a question.

http://www.diystompboxes.com/pedals/schems/msim.jpg
This one is a direct rip from a Marshall amp with an XLR output. I do have one concern though, only the first two op amp stages have the non-inverting input referenced to VB (bias voltage, presumably). Why is that? Is it because the second two stages are not AC coupled to the first two stages? Does the bias voltage carry through? I would think the voltage would be off because the first two stages have gain. Will the design work as-is? I also believe I should add a DC blocking cap on the output of the last stage, correct?



http://www.schematicheaven.com/marshallamps/jtm30_30w_jtm310.pdf
Here's a link to the full schematic of a Marshall amp that uses this circuit. I'm not finding any answers there because it uses a bipolar supply rather than a bias voltage like the first link I posted. The section of interest is IC3 (b - e) which is on the right side of the first page about half-way down. You can see right above it that the IC is powered by two voltage regulators (78L15 for positive supply and 79L15 for negative).

If anyone is curious, here's another design that is even simpler (using only one dual op amp) I don't have a schematic for it so I'll have to create one from the PCB design. http://www.radmer.dk/billeder/amz-lay.gif This seems to be an AMZ design but the schematic is no longer on www.muzique.com

[edit: Oops... I forgot to include a link to the real marshall schematic ]

[teemuk]

Is it because the second two stages are not AC coupled to the first two stages? Does the bias voltage carry through?

Bingo!

For the same reason it also needs a coupling cap to the output. For example, if that thing is powered from a 9-volt battery the signal's "zero point" will be the Vbias (4.5V) and this potential will also appear in the output.

The gain can screw the DC offset but I think the chance for that is quite remote with only a couple of stages that are also AC coupled from each other. And the two final stages don't have gain at all. As you can see, they are mostly voltage followers with a gain of unity, except at high frequencies at which they begin to acts as second-order low-pass filters.

--

I see if I can dig up the schematic to one of Gallien-Krueger's cabinet simulators. It was a really simple one too, used just a single opamp as well. But those simple ones aren't neccessarily the best ones or very realistic, though.

[armstrom]

Thanks so much for the quick reply. I suspected this was the case but I wasn't 100% sure (The condor cab sim from ROG uses a similar design with only the first two stages biased directly).

What size cap would you use on the output? Presumably something large enough not to affect the frequency response maybe a 0.47 uF? Or bigger... 1.0uF?

I would love to take a look at those schematics. I'm trying to come up with something super easy to build but also something that I can squeeze into an empty space in my preamp box. I built a little preamp for my friend to record with. He's using software amp sims currently but I would like to filter out some of the highs in the box so maybe it could be used with a PA. I hear what you're saying though about simple designs not being realistic. There's no way a couple of filters are going to reproduce the complex frequency response of a guitar speaker in a cabinet. My hope is that I can get "close enough" with the overall frequency response that the PA speakers will sound somewhat convincing to the ear (rather than just sounding tinny like a guitar plugged directly into a PA does)

[teemuk]

Here (attachment). In case you wonder, VCV is G-K's acronym for "Vintage Cabinet Voicing". This is so far the most simple circuit I've seen that is able to provide a quite decent simulation. Personally, I might pick something like the Marshall sim instead but if you need something extremely minimal then this one would be a considerable option.

[armstrom]

wow, yeah that is simple! might just do the trick. I'm pretty sure I have all the parts here to throw one together and try it out.

[armstrom]

Here's another resource with several cab sims ranging from simple to complex.
http://www.herby.kielce.pl/~piter/hexenew2/ie.php?c=cabsims_e&l=diy_e

Edit: Oops.. I just saw your name in the article, I guess you translated it I'll leave the link here since it's pertinent to the discussion though.

-Matt

[Gatopanzon]

Here (attachment). In case you wonder, VCV is G-K's acronym for "Vintage Cabinet Voicing". This is so far the most simple circuit I've seen that is able to provide a quite decent simulation. Personally, I might pick something like the Marshall sim instead but if you need something extremely minimal then this one would be a considerable option.

i register because of this gif, thanks!!

[phatt]

I highly recommend the *Ed Rembold* circuit you posted,,It delivers the big cut below 100hZ, I built several and they went like hot cakes. <3)

Having messed around with several cabsim circuits,, :loco
My Rule of thumb, find a circuit with the most cut below 100Hz and above 3~4 kHz and it will be close enough. With no bottom end roll off it will turn to mud when cranked up.
For recording I found the best result came from inserting cabsim *After* my old Quadraverb GT unit,, not before it. YMMV

You may find it useful to still have some kind of tone control circuit along with cab sims as there will always be differences between your gear and others.

You can play around with the 33k resistor value,, try 22k for a brighter edge but don't overdo it. :trouble

Oh don't forget the input Z is lowish so direct input from passive pu's will be less than ideal.
*IF?* I where to build Ed's circuit again I'd put a high Z buffer in front and another after it,, guarantees it will match to almost anything but that means another chip.
Cheers,, Phil.