ilovefuzz.com

I've mentioned this to a few of you guys before, but my favorite ring mod has always been the EHX Frequency Analyzer. Well I realized a few weeks ago that I'm without a ring mod and I've tried at least a half dozen of the most popular ring mods around here. They're all good pedals, but the only one that really yanked my crank sound-wise was the FA. But the power requirements and the lack of controls are the only things I don't like about it. So I've decided to clone the bugger, add some more controls and fix the power section.

The schematic is pretty easy to find, and the circuit is not a complex one. The component count is pretty small so I figure this shouldn't be all that difficult. I've ordered all the parts and they're mostly all here now. The only snag is that a couple of the chips are obsolete.

The waveform generator for the carrier signal is an ICL8038CCPD, which is marked as obsolete with no recommended replacements. It's actually a pretty powerful chip that has several different waveforms and good control of frequency and duty cycle. I was able to find a source on eBay, so I ordered a half dozen of them. I'm sure I'll find good use for them at some point.

The multiplier/divider is an AD553JH, which is also obsolete. I was able to find a few of these on eBay as well, but they were so expensive (like 75 bucks a piece ) that I decided "fuck that." There are no pin-for-pin drop-in replacements, but there are chips that are functional equivalents. I was able to score a couple AD532JD's which are pretty damned similar, and it will actually allow me to simplify the circuit, because it's internally trimmed, which means I won't need the calibration trimpots.

The only other issue is figuring out the power section. The standard EHX FA required a 40VDC adapter, which is simply ridiculous. The reason it needs 40V is because the waveform generator requires +15V and the multiplier requires bipolar +-15V. I really want this thing to run on a standard 9V supply. I found that the MAX1044 charge pump can double your positive voltage, and can also flip the polarity and give you -9V. Actually, it can give you both +18V and -9V at the same time from one chip. And then if you daisy chain another MAX1044 onto the -9V, you can double that to -18V. I modeled it in LTSpice first to make sure, and then threw the power section on my beavis board last night. Works exactly like it's supposed to, I actually measured +20V and -19V from the circuit, all from one Boss PSA.

Over the next few nights, I'll hopefully breadboard the rest of the circuit. I'll update this thread with my progress, and hopefully some clips. Wish me luck!

bigchiefbc wrote: Wish me luck!

Yes, good luck.

Seems like a cool project.

Nice project, you might wanna go for the lt1054 rather than the 1044. It can handle a higher input voltage (+15 vs +10 i think) which is good for all those not using a well regulated supply. 1044s can be prone to die from even small voltage spikes on the supply rail. Also that waveform generator, in the datasheet it says it'll work on a bipolar +- 9V supply so you don't really need to go to the trouble of using 2 charge pumps. The AD633 will work on +-9V as well which is also another good option.

mysteriousj wrote:Nice project, you might wanna go for the lt1054 rather than the 1044. It can handle a higher input voltage (+15 vs +10 i think) which is good for all those not using a well regulated supply. 1044s can be prone to die from even small voltage spikes on the supply rail. Also that waveform generator, in the datasheet it says it'll work on a bipolar +- 9V supply so you don't really need to go to the trouble of using 2 charge pumps. The AD633 will work on +-9V as well which is also another good option.

Good to know about the 1054 vs the 1044, I'll have to look into that.

You are correct about the fact that I could have used bipolar 9V for the 8038, good call. As for using the AD633 to eliminate the need for the second charge pump - I may give that a shot, we'll see. I already have the AD532s, so I'll at least do my initial tests with that. But thanks for the heads up on that, I hadn't seen that chip yet. Everywhere I read said that the 532 was the closest replacement for the 533, so that's really all I looked at.



As an update, I only had a little while to play around tonight, but I got the carrier section hooked up and getting a good waveform, sine, triangle and square. I spent a little while tweaking component values to get the frequency range I wanted. Going good so far. Hopefully I'll have time to put the rest of the circuit together in the next few days.

oh man gooood luck (:

Awww shit.

If this works out i may ask u to build me one. Love the musical-ness and dirtyness of the freq analyzer, not something found in alot of ring mods besides the randy's revenge, which still sounds too cold in comparison

YOU'RE A MADMAN, I SAY

YAY, finally an update!!

Finally got the whole thing breadboarded tonight. It works...almost. It is passing signal and is definitely ring-modding, but the signal is way too fuzzy and sputtery, even when the blend is fully clean. The carrier oscillator is perfectly clean and sweet sounding. So I gotta do some debugging I didn't have time for tonight. But, it's definitely progress!

Is there any chance of you making this into a pcb, or is there anyone with a ring-mod PCB out there I could buy, and Logan ring mods out there left?

bigchiefbc wrote:YAY, finally an update!!

Finally got the whole thing breadboarded tonight. It works...almost. It is passing signal and is definitely ring-modding, but the signal is way too fuzzy and sputtery, even when the blend is fully clean. The carrier oscillator is perfectly clean and sweet sounding. So I gotta do some debugging I didn't have time for tonight. But, it's definitely progress!

The4455 wrote:Is there any chance of you making this into a pcb, or is there anyone with a ring-mod PCB out there I could buy, and Logan ring mods out there left?

if you want a free pcb for the CA ring mod send me a PM. I have given a bunch out so i even have reasonable documentation at this point (;. It is not as smooth sounding as this will probably be though, its more of a fuzz/ring mod.

The4455 wrote:Is there any chance of you making this into a pcb, or is there anyone with a ring-mod PCB out there I could buy, and Logan ring mods out there left?

Assuming I get this thing debugged and sounding perfect, then yes I'll definitely be getting a PCB made for it, I wouldn't want to have to make a circuit like this on perf or vero. But it may be a ways off, my free time consists of like one evening every couple weeks.

Yeah multi_s hooked me up with some ring mod boards I can't wait to put them together and make some robot noises.

bigchiefbc wrote:

The4455 wrote:Is there any chance of you making this into a pcb, or is there anyone with a ring-mod PCB out there I could buy, and Logan ring mods out there left?

Assuming I get this thing debugged and sounding perfect, then yes I'll definitely be getting a PCB made for it, I wouldn't want to have to make a circuit like this on perf or vero. But it may be a ways off, my free time consists of like one evening every couple weeks.

definitely feel your pain on the free time. but would also definitely be interested in a pcb if it ever comes to be.

Please post a clip!

Multi's CA ring mod is one of my favorite effects, btw.

ya i like the sound of it too but in the back of my mind im always like 'this isnt _true_ ring modulation' which may or may not create a more desirable sound. i used the frequency analyzer and its pretty nice. different texture. obviously id be using both. (:

i was farting around with some dsp based ones but eventually stopped due to time constraints from school and other projects. i sort of like looping sequences that come out of ring mods and trying to improv over them, its an interesting effect, creates some unexpected results.