I want to build a multipedal. My issue is, I hear this "heterodyning" problem with putting multiple modulation effects in one enclosure.
1. What is it?
2. How can I prevent it?
3. The effects I plan on putting into one enclosure are a flanger (current lover), and a nom nom (phaser), and if possible an analog delay.
Thank you for any and all help everyone,
Brandon
You should put a reverb in there too.
Sent from my iPhone using Tapatalk
Deal. But how do stop the heterodyning that I've heard of? Is it a mix of signals, or a whine, buzz, or something else??
This is from the DIYSB FAQ
QuoteI am picking up radio stations, how can I stop this?
From Eric Hensel:
A 47pf cap to,ground, after the input cap will probably do it --you can adjust this up or down --higher values will start to cut treble. use the lowest value that works.
From Zachary Vex:
mount the pc board as close as you can physically to the metal box. this will serve as a ground plane to reduce the heterodyning that can pick up rf. don't let the input and output wires cross... make sure the layout doesn't allow input and output connections to get too close to each other. if you can, solder in ground connections around (near) the sensitive input circuitry. if rf is being picked up by rectification, these fixes won't help, but if it is being picked up due to beat frequencies generated by heterodyning (internal rf oscillation in the circuit beating against rf from radio stations) you can get rid of it through the use of careful layout and grounding to reduce oscillation.
you might try using miniature shielded cable inside of the enclosure for the input to the board too.
Cody
I've generally only heard of this being a problem with multiple circuits using charge pump chips, OR with multiple circuits with different clock signals for each.
Juansolo has the most experience with this.
Quote from: sikmunky on March 01, 2015, 03:05:01 AM
Deal. But how do stop the heterodyning that I've heard of? Is it a mix of signals, or a whine, buzz, or something else??
yeah, I should be said I had no clue, but I got excited about the build.
Sent from my iPhone using Tapatalk
Fair enough. Lol. I had read juan solo's page regarding heterodyning and then Rej's (grind customs) build with a chorus, flanger, and phaser in one enclosure, and thought, I have no idea what this is (regarding g the heterodyning thing). But I want to have my cake and eat it to. So the least amount of patch cablez, real estate, and hum, the better
It's a clock thing, and as Alan says, it's most common with multiple charge pumps. I now have a one charge pump rule (and run multiple things from it) for multis because I've had issues a few times in the past.
But it's to to with the frequency the clocks run at. So in theory, multiple clocks from modulation can cause it, but in reality it's going to be rare as the clocks are running so slowly and it's unlikely for them to all be running at around the same speed. So it's pretty safe.
The other big problem is with clock noise I've found. The CE-2 is a big offender in this one. The wires on the pots act like a transmitter for it (it seems to be a CE-2 specific issue...). You want them as short as possible, or shielded. You just need to be careful where you run wires.
All the above is why I always fit delays without modulation to multis and try and keep clocks to a minimum. That said, the Alice 2.5 I'm currently building has 3 in it all next to each other, so what do I know ;)
For RFI try a 3K3 resistor on the input and a 470pf capacitor to ground after it.
When diagnosing a problem like this or simply using shielded cable in your build to avoid it, do you ground both sides of the cable or just one like in the input of a high gain amp? My only experience with a similar issue was a JCM800 build, and grounding both sides caused its own problem. Just asking out of curiosity.
Shielded cables, just ground one end. With the CE-2, I ended up putting a metal plate between the CE-2 pots and the Delay (that was picking up the clock ticks) then connecting the plate to ground.
Good to know. So I'm thinking the delay in the same enclosure would be a disaster waiting g to happen. But after looking at the build docs for the flanger and phaser, it looks as the flanger is the only one with a clock? Would I need to do anything out of the norm for wiring? Would I want to throw a 3prr board and charge pump on for increased voltage/headroom? Is the dpdt filter switch on the current lover a on-on switch, or on-off-on switch?
Once again, thanks a bunch guys,
Brandon
I think the Electric Mistress might be the issue with this one... I have a feeling you'll have trouble with that just based on how it behaves when tested out of a box (it really doesn't like it). Not so much heterodyning, more the fact that it needs to be inside a box before you set it up, so it really likes having the shielding... That's the one I'd be wary of (it wouldn't be my choice for a multi).
The delay hasn't got a clock but the phaser has. The only trouble I've had with an analogue delay in a multi was fitting it in with a CE-2. That was the first time I had two charge pumps heterodyning. So I tried running the pair of them from one charge pump, which didn't have enough grunt. In the end I went with one effect at 9v and the other at 15v
As I say, the slower speed clocks are unlikely to cause any issues. It's not impossible, it's just unlikely.
Thanx a bunch for all the help guys. As much as I would like to make an epic multi build, I think I'd be getting in over my head. Hopefully once my component order gets in, I'll be able to finish and box em up. Until then, I'll wait and try to finish the final wiring of my timmy, rat, russian muff multi.
I tried building an MXR Mico Flanger and an MXR Micro Chorus both into a 1590BB. I was successful with the builds however, I was VERY unsuccessful with heterodyning. Both of these circuits do not use charge pumps however, everything I tried failed to eliminate the clock issues. Proper shielding, star grounding, RF separation, caps, In/Out wire run separation... I tried everything.
In the end, I had to put each circuit into its own 1590B.
Of course, these were single-sided etched PCBs with no ground pours so fabbed boards might help. Also, the smaller enclose did not allow for a great deal of circuit separation.
You might get lucky and I hope you do :-\
I was planning a Rub a Dub reverb with a Smoothie on the wet signal soon - is this something that could be expected with those two combined?
Quote from: juansolo on March 02, 2015, 09:12:44 AM
I think the Electric Mistress might be the issue with this one... I have a feeling you'll have trouble with that just based on how it behaves when tested out of a box (it really doesn't like it). Not so much heterodyning, more the fact that it needs to be inside a box before you set it up, so it really likes having the shielding... That's the one I'd be wary of (it wouldn't be my choice for a multi).
The delay hasn't got a clock but the phaser has. The only trouble I've had with an analogue delay in a multi was fitting it in with a CE-2. That was the first time I had two charge pumps heterodyning. So I tried running the pair of them from one charge pump, which didn't have enough grunt. In the end I went with one effect at 9v and the other at 15v
As I say, the slower speed clocks are unlikely to cause any issues. It's not impossible, it's just unlikely.
Just to clarify, Analog Delay
does have a clock, PT2399 delays too but it's not usually an issue for us. The phaser doesn't (only PWM phasers do) it has an LFO which is a source for tick but not heterodyning.
With 2 charge pumps without very good layout they will interact as they're at the same frequency, which is why the LT1054 has pins to adjust the clock so you could use two but you'd then have to worry about the clock of the flanger, chorus or delay reaching the same frequency and interacting.
So with a flanger and delay you probably wont have issues as the flanger clock is likely a lot higher than a delay clock, but a delay and a chorus work in the same range so are more of a cause for concern. A charge pump is usually in the 25-45kHz range clock wise so that's a problem for delay and chorus, but both can work in one build, it's really down to power, layout and luck.
Indeed, and nailed on.
Quote from: mcallisterra on March 03, 2015, 03:52:55 AM
I was planning a Rub a Dub reverb with a Smoothie on the wet signal soon - is this something that could be expected with those two combined?
pure awesomeness!!!!!
Sent from my iPhone using Tapatalk