I wanted to jump in here and say that in the past year, I've repaired 4 Echoflanger/PolyChoruses. I used the old factory schematic and the calibration document Ron posted, as well as the board re-draw that Nelson did long ago. This is a headache of a pedal to work on, but they really sound great. I was thinking of trying to come up with a rejuvenated clone with current production BBDs, but someone has beat me to the punch - good job! The idea of using two MN3007 is interesting. I was thinking to use an MN3007 for the 1024 stage, but then have two MN3009 in series for the 512 stage, maintaining the single clock frequencies of the original... I'm still thinking I might play with that idea just to see how it works.
Regarding the documentation for the Dandy-Horse, I'm surprised to see that it doesn't insist on using an oscilloscope for calibration. I would consider this mandatory for getting the trimmers set correctly. In reference to the original EHX documentation, you set the bias trimmers to give symmetrical clipping - I do not see how this could possibly be judged by ear. Also, a sine wave oscillator should be considered mandatory for the signal source at this point, because there's no way you will get a consistent steady signal from a guitar to accurately be able to judge the settings. Same goes for setting the level trimmers: you must be able to measure accurately that the the straight and delayed levels are exactly equal in Filter mode and delay is exactly 10% higher in Slapback mode. This is easy with an oscilloscope and virtually impossible by ear. And the "10% higher" might seem like something hard to see, but EHX's instructions say to set it so that the straight level is 1v P-P, so the 10% higher level is an easy to calculate 1,1v P-P, which even my old analog scope could show when zoomed in enough. Another very important thing, that isn't mentioned either in the Dandy-Horse or the EHX instructions, is that the Blend switch MUST be set in the OFF position when adjusting the levels. Having it on will give you false readings of the effected signal level that will throw off the whole procedure.
Setting the clock cancellation trimmer by ear is also an impossible task, because it is at a frequency beyond audibility. The whine you hear isn't the entirety of the clock, and when it "disappears" likely something is still there. And that inaudible clock will be modulating your signal and possibly causing distortion. And again, this is easy to do with a scope and you can keep zooming in on the wave until you get it trimmed down to as close to completely gone as the trimmer will allow.
Now, despite that rant about the need for a scope, the one thing that I think can and should be trimmed by ear is the Feedback. I think the behavior of the feedback is something that can come down to individual taste. And this is also when you need to plug a guitar in to set it, because the feedback sensitivity should be set based on how you play. I think one of the most rewarding tones you can get from this circuit is when you have the feedback just on the edge of oscillation, where certain notes will jump out and start to go wild, but then calm back down again as you continue playing. Usually I set the trimmer for full on oscillation when Feedback is maxed, which usually gives good room to dial in the on-the-edge setting. But maybe someone wants to be more conservative with it and not allow oscillation - that's up to personal taste. I don't think EHX's instructions are very good in this regard, because following their instructions seems to always result in a lack of oscillation. and I don't think you can really set it in a musical way with a constant sine wave as the source.
Related to this discussion. I have a 90's-00's era Stereo PolyChorus and can attest that it sounds absolutely nothing like the 70's era Echoflanger/Polychorus. Nothing about it sounds the same at all, even the behavior of the LFO seems different, and the delay times I'm getting in Chorus/Double Track seem at least twice as long as what the Echoflanger can do. Mine might have a misadjusted clock? I haven't tried digging into it yet to see, but that likely is more caused by the fact that the BBD has more stages in that mode (MN3008). And Flange mode has half as many stages as the Echoflanger? Anyway, I'm not saying the Stereo PolyChorus sounds bad at all - it sounds great! And the extra long delay of the Double Track mode makes it a legit analog delay pedal if you're not wanting super long echoes. They shoulda gone with the alternative "Slap Back" name because that's what it sounds like, much too long for a fake Double Track sound. I was initially thinking about trying to mod this to be more like an Echoflanger, but now that I've played with it some more, I really like it for what it is.
Thanks to everyone who has contributed info to this thread and these projects, and special thanks from me to Ron and Nelson for providing the documentation which has been so helpful in my repairs of these things.
EHX - Echoflanger [schematic]
-
soulsonic
- Old Solderhand
Information
"Analog electronics in music is dead. Analog effects pedal design is a dead art." - Fran
-
soulsonic
- Old Solderhand
Information
And some notes from me that might be helpful to anyone else working on an old Echoflanger:
As mentioned in this thread already, the mode rotary switch will often go bad. I found it impossible to find a replacement with the same pin layout as the original. I was able to make it work with a standard Lorlin switch with solder lugs; I bent the lugs out and attached wires to them to get them connected to the correct pads. I put heatshrink over the lugs so they wouldn't short against anything else on the board. There was enough room for it to fit between the board and the case still. I had to cut the shaft to be a D shape with the correct orientation for the original knob. I used the plastic version of this switch, so that wasn't too hard.
You see in the photos at the beginning of the thread that there is this weird paper strip going across the board. That is making the connections of the Ground, +15v and +7,5v throughout the circuit. I had this actually go bad somehow on one of the units I repaired many years ago. The only solution was to pull it out and replace it with individual wires making the buss connections. On a unit I just repaired, this was actually a set of pins connected with wire wrap. I've probably seen 6 different specimens in my life and this was the only one done this way instead of using the paper-covered buss strips.
The IC sockets are a common source of problems. If you have a fault traced down to a specific IC, try pulling it out and reinserting to see if that solves the problem. If it helps, you should then replace the socket. I think replacing IC sockets in old EHX pedals is right up there with replacing all the electrolytic caps as far as being almost mandatory at this point.
The original trimmers were often not great quality, and I've seen problems caused by them failing. There will often be a mix of different brand trimmers used, and the black plastic Piher ones are usually not a problem, but there's these open frame ones (you see this kind in the photo at the beginning of the post) that are very flimsy and I have often had a hard time getting a screwdriver to properly slot into it from behind through the hole in the board - my Bournes trimmer tool won't fit at all, though it does fit the Piher trimmers. Consider the trimmers suspect when you have problems traced to an area where one resides. I've encountered ones where the wipers were worn out and ones where the phenolic board with the resistance track on it was cracked. I've had them almost totally fall apart after attempting to adjust.
If you're going to be doing alot of work on the board, like replacing all the old electrolytic caps, it is very wise to go ahead and unsolder all the flying leads from it. They are only going to break off anyway, I guarantee it. EHX's use of solid core wire doesn't do well here.
Clean that Blend switch well, even though no signal is going through it, you can still get intermittent signal issues if it is dirty.
I've often tried changing out the 4053 chips, thinking that they may be fried and the source of the issue I was having, but this has hardly ever actually solved the problem. Maybe it was the case once... Don't rush to thinking they're dead.
In addition to the old electrolytic caps, replace those tantalums too! I don't trust tantalums, especially old ones. They can fail with intermittent shorts and cause all kinds of mayhem. Replace the two 220nF ones at the compandor with film ones. I'm sure the only reason they used tantalum for those back in the day was the issue of size. I actually replaced the two 1uF ones with MLCCs and it wasn't any the worse for it.
If the unit doesn't have a 3 prong grounded power cord, bear in mind that you will need to connect the chassis ground to your scope's ground when making measurements during calibration. I was working on two units, one with grounded power cord and one without, and got confused why the scope was showing such a dirty modulated clock signal on the one with the 2 prong cord. Finally realizing my mistake and connecting the scope ground to the chassis cleared up that bonehead moment. Hadn't even noticed that the 3 prong one had the ground connection being made through the shared safety ground. Incidentally, it might be wise to install a 3 prong cord if it doesn't have one. I connect the ground lug to one of the screws holding the power switch. You can re-use the original strain relief if you carve it out to make space for the thicker cable.
If you think you might have a dead BBD chip, my heart goes out to you... But a quick way you can verify this is to swap the two chips and see if the problem "follows" it to the other mode. If Chorus isn't sounding, swap chips and see if now Flange isn't. Unless of course BOTH chips are dead, then you may as well throw it off a bridge.
When trimming out the clock signal cancellation, keep zooming in on the vertical axis of the scope. It may look like it is gone, but I almost guarantee if you keep zooming in, you will likely still see something there. The input signal should be disconnected when doing this. It is often impossible to trim it down to absolute zero, but you should be able to get very close.
I mentioned this already above, but I feel I need to reiterate it - you absolutely must have the Blend switch OFF when calibrating levels. And an oscilloscope + sine wave oscillator should be considered mandatory.
As mentioned in this thread already, the mode rotary switch will often go bad. I found it impossible to find a replacement with the same pin layout as the original. I was able to make it work with a standard Lorlin switch with solder lugs; I bent the lugs out and attached wires to them to get them connected to the correct pads. I put heatshrink over the lugs so they wouldn't short against anything else on the board. There was enough room for it to fit between the board and the case still. I had to cut the shaft to be a D shape with the correct orientation for the original knob. I used the plastic version of this switch, so that wasn't too hard.
You see in the photos at the beginning of the thread that there is this weird paper strip going across the board. That is making the connections of the Ground, +15v and +7,5v throughout the circuit. I had this actually go bad somehow on one of the units I repaired many years ago. The only solution was to pull it out and replace it with individual wires making the buss connections. On a unit I just repaired, this was actually a set of pins connected with wire wrap. I've probably seen 6 different specimens in my life and this was the only one done this way instead of using the paper-covered buss strips.
The IC sockets are a common source of problems. If you have a fault traced down to a specific IC, try pulling it out and reinserting to see if that solves the problem. If it helps, you should then replace the socket. I think replacing IC sockets in old EHX pedals is right up there with replacing all the electrolytic caps as far as being almost mandatory at this point.
The original trimmers were often not great quality, and I've seen problems caused by them failing. There will often be a mix of different brand trimmers used, and the black plastic Piher ones are usually not a problem, but there's these open frame ones (you see this kind in the photo at the beginning of the post) that are very flimsy and I have often had a hard time getting a screwdriver to properly slot into it from behind through the hole in the board - my Bournes trimmer tool won't fit at all, though it does fit the Piher trimmers. Consider the trimmers suspect when you have problems traced to an area where one resides. I've encountered ones where the wipers were worn out and ones where the phenolic board with the resistance track on it was cracked. I've had them almost totally fall apart after attempting to adjust.
If you're going to be doing alot of work on the board, like replacing all the old electrolytic caps, it is very wise to go ahead and unsolder all the flying leads from it. They are only going to break off anyway, I guarantee it. EHX's use of solid core wire doesn't do well here.
Clean that Blend switch well, even though no signal is going through it, you can still get intermittent signal issues if it is dirty.
I've often tried changing out the 4053 chips, thinking that they may be fried and the source of the issue I was having, but this has hardly ever actually solved the problem. Maybe it was the case once... Don't rush to thinking they're dead.
In addition to the old electrolytic caps, replace those tantalums too! I don't trust tantalums, especially old ones. They can fail with intermittent shorts and cause all kinds of mayhem. Replace the two 220nF ones at the compandor with film ones. I'm sure the only reason they used tantalum for those back in the day was the issue of size. I actually replaced the two 1uF ones with MLCCs and it wasn't any the worse for it.
If the unit doesn't have a 3 prong grounded power cord, bear in mind that you will need to connect the chassis ground to your scope's ground when making measurements during calibration. I was working on two units, one with grounded power cord and one without, and got confused why the scope was showing such a dirty modulated clock signal on the one with the 2 prong cord. Finally realizing my mistake and connecting the scope ground to the chassis cleared up that bonehead moment. Hadn't even noticed that the 3 prong one had the ground connection being made through the shared safety ground. Incidentally, it might be wise to install a 3 prong cord if it doesn't have one. I connect the ground lug to one of the screws holding the power switch. You can re-use the original strain relief if you carve it out to make space for the thicker cable.
If you think you might have a dead BBD chip, my heart goes out to you... But a quick way you can verify this is to swap the two chips and see if the problem "follows" it to the other mode. If Chorus isn't sounding, swap chips and see if now Flange isn't. Unless of course BOTH chips are dead, then you may as well throw it off a bridge.
When trimming out the clock signal cancellation, keep zooming in on the vertical axis of the scope. It may look like it is gone, but I almost guarantee if you keep zooming in, you will likely still see something there. The input signal should be disconnected when doing this. It is often impossible to trim it down to absolute zero, but you should be able to get very close.
I mentioned this already above, but I feel I need to reiterate it - you absolutely must have the Blend switch OFF when calibrating levels. And an oscilloscope + sine wave oscillator should be considered mandatory.
"Analog electronics in music is dead. Analog effects pedal design is a dead art." - Fran
-
Dirk_Hendrik
- Old Solderhand
Information
Great writeups Martin. Thanks.
Allow me to add to this for anyone dealing with issues like these
Especially when working with BBD's or other hard to source obsolte parts, do work ESD safe. If you don't know what ESD safe is, look it up and please, do not consider it some secondary thing just because it adds another hickup in the long and difficult repair process. It may be he difference between a quick repair and a way longer and more expensive one.
Allow me to add to this for anyone dealing with issues like these
Especially when working with BBD's or other hard to source obsolte parts, do work ESD safe. If you don't know what ESD safe is, look it up and please, do not consider it some secondary thing just because it adds another hickup in the long and difficult repair process. It may be he difference between a quick repair and a way longer and more expensive one.
-
uncleboko
- Cap Cooler
Very wise words - memories of when I lost a SAD4096 back in the 1980s!Dirk_Hendrik wrote: ↑11 Jan 2025, 14:59 Great writeups Martin. Thanks.
Allow me to add to this for anyone dealing with issues like these
Especially when working with BBD's or other hard to source obsolte parts, do work ESD safe. If you don't know what ESD safe is, look it up and please, do not consider it some secondary thing just because it adds another hickup in the long and difficult repair process. It may be he difference between a quick repair and a way longer and more expensive one.