KMD Analog Delay Question/Tone control

[trovador]

Hi Folks, this is my first post!

I have a cool, kinda rare, KMD analog delay from the 80's made in Japan for Kaman/Ovation. I don't know who designed or manufactured it but it is pretty much a Boss copy. Very close in design to the DM2/3 but it differs in that it has an additional 'stereo' output (Not 'direct out' as the DM-3) and has a 4th knob for tone control. The tone control cuts the high frequencies of the signal, not just the delay, and continues to affect the signal even if in 'off' position. My question is: Is it possible to modify it so that the tone affects only the repeats? so that when off it won't affect the tone of the dry signal? Thanks.

Here's the pedal:



Inside shot:



I don't know how well it will show, but here's a copy of the schematic:

[Enzo]

The tone control is right at the input stage, so it is going to affect everything that goes through the pedal. You'd have to remove it, and create a new tone control, probably just left of the JFET by the repeats control.

[trovador]

Enzo,

So, I'm assuming it's possible based on your response. What's the degree of difficulty for such a project?

Jose

[Enzo]

Possible? Sure, anything is possible. I have no idea how difficult it would be for you. I was just pointing out the reason it works like it does is the place in teh circuit. It would have to be moved. Where I mentioned seems like the best spot. You could try a simple tone control like on a guitar... might work.

[trovador]

Thanks. I'm not an expert, but I can do minor mods if directed.

So if I hear you correct, forget about the tone circuit in the unit (or disable it?), if I want a tone shaping for the delays I should add a new tone control where you suggested (just left of the JFET by the repeats control)

[Enzo]

Whatever you might do to the existing tone circuit, it will affect both FX and dry signals, so there is no modification that will achieve what you want at that spot. You could either remove it, or leave it set to sound best and then not used. n Then you would need to add a new circuit in the FX path, but before that path mixes back with the dry. SO experiment. I figure the point I mentioned would be about the latest point, but earlier positions might work too.

[trovador]

Whatever you might do to the existing tone circuit, it will affect both FX and dry signals, so there is no modification that will achieve what you want at that spot. You could either remove it, or leave it set to sound best and then not used.

Gotcha. That's in fact what I do, I just leave the tone pot where sounds best. If I actually leave it all the way to the right, it doesn't affect dry the signal when disengaged (to my ears, at least) and functions as a Boss DM-3 would with the regular 3 controls. It's not a bad delay at all.

Then you would need to add a new circuit in the FX path, but before that path mixes back with the dry.

Yep, that's the challenge! I don't know if I'm completely up for it, yet. Maybe if I can find another cheap one and experiment...

Thanks for the suggestions!

[g1]

Another option that would deal with the issue is true bypass switching. However, the design of the switching mechanism means true bypass would be a fairly difficult mod.
Perhaps some kind of A/B pedal would allow complete bypass of the delay pedal.

[trovador]

Perhaps some kind of A/B pedal would allow complete bypass of the delay pedal.

yes, I think this is the easiest alternative, building true bypass looper pedal

[Enzo]

That would work, but you lose the ability to affect only one channel with the tone control.

[Mark Hammer]

The tone control actually doesn't cut treble. Rather it adds varying degrees of treble boost. Because there is the usual 8 poles of lowpass applied to the delay signal, to keep noise out of the output, you notice the treble boost on the clean signal more than the delay.

Personally, I think it is a useful control to leave in, but that's just me.

You probably don't want to increase the treble on the delay-only, since that would increase noise. However, it is possible to remove treble from the delay signal, closer to the output stage. When you say "remove it from the repeats", that can mean two things. It can mean changing the tone of the entire delay signal, OR it can mean changing the tone of each repeat after the first one. Both are feasible.

I prefer changing the tone of the additional repeats. This allows you to crank the Repeat control, without having to worry about runaway feedback and sound degradation. It also sounds more natural. In the real world, echoes that repeat have bounced off several reflecting surfaces, but each reflection has less treble, simply because the reflecting surfaces are imperfect, and sound energy is lost through air as well.

The simplest mod for this would be to identify the resistor and cap that come off the wiper of the Repeat control. I wish to heck I could see the value of the resistor, but I can't tell what it says. However, whatever it's value is, you will want to replace that with a pair of resistors that add up to the same value. So, if it was a 22k unit, you would replace that with a 10k and 12k in series. If 10k, perhaps a pair of 5k1 units, or a 5k1 and 4k7.

The junction of those two replacement resistors will be our point of intervention. What you'll do is connect a cap from that junction to ground. The objective is to select a cap value that will produce a rolloff around 1khz. The formula is Freq = 1 / [2 * pi * R* C], where R is in megohms, and C is in microfarads. If the first resistor was 10k, then a suitable cap value would be .015uf. Not exactly 1khz (1061hz, using the formula, and presuming the parts measure exactly those values), but close enough.

Why this value? Well, the filtering on the unit will probably be set somewhere around 3-4khz. Given that the additional repeat filtering we are installing is only a puny 1-pole filter. It will only remove a little bit of treble, and remove a little more, and a little more with each repeat. By the time the signal has repeated 3 or 4 times, it will have gotten somewhat degraded, and grittier. BUt since that grittiness is in the upper registers, the progressive lowpass filtering we've added will trim it off as it accumulates. The fundamental note in each repeat will still be quite audible, but each repeat will appear to move further into the background, because of the progressive treble loss. To my mind, this makes for a less "cluttered-sounding" delay.

You can install this as a fixed change, or you can install a little toggle if you have room, and use the toggle to connect and disconnect, the cap from ground, to engage and disengage the effect.

If you try this, let me know if you like it.

[trovador]

I wish to heck I could see the value of the resistor, but I can't tell what it says.

Does this help?

[Mark Hammer]

A great deal, actually. Split that 27k resistor into a 15k and 12k in series, with the 15k "first". Run a .01uf cap from the junction to ground. If you find that doesn't cut enough treble on successive repeats, use .012uf.

[trovador]

I experimented a little over the weekend and first, disabled the tone pot so that it doesn't affect the signal when off. After playing around with it, I discovered that, I do like the tone where it is. It does add something neat, so I re-soldered it back to stock. Now, I would still like play with the repeats. I did try Enzo's suggestion different ways, but nothing that sounded good to my ears, perhaps not doing it right... I'm intrigued by Mark's suggestion, especially if it could have a switch to give the options but I'm afraid it's a little too advanced for me...

[Audiotexan]

Does this help?

<photobucket img snipped>

Can you just post the whole schem here please trovador?
-strike the above-

found a better/readable shot that he'd already included in his photobucket collection

http://smg.photobucket.com/user/mase...5ab1e.png.html