Memory Man Calibration?

[Drewmeyer]

So I just pulled the trigger on an old non-deluxe Memory Man:

So far I'm loving the pedal, except for a few things. First, the repeats seem to distort incredibly quickly (although I've never played any other memory man for comparison) and there seems to be a high pitched whine kind of noise over top of the delayed notes. I looked inside and noticed there were about six or so trimpots on the board and after some research discovered that some people suggest Memory Mans need to be re-calibrated after certain periods of time when they begin showing certain symptoms and I noticed my pedal is showing the same symptoms as those mentioned. So my question is, does it sound like I should calibrate the pedal? If so, who would you recommend that can do such things?

[Mark Hammer]

Yes.

The distortion, as described, sounds like the bias trimpot is off.  When the bias is set right, the wet signal is relatively clean.  When set a little too high or low, you can hear the delay but its distorted.  When set higher or lower than that, you don't hear any delay at all.

The audible whine sounds like it could be due to one or both of the following: need to adjust the clock frequency upwards a bit, need to tweak the balance trimmer on the output of the BBD so that maximum clock signal cancellation is achieved.  My own memory may be playing tricks on me, but if I recall the late 70's with any accuracy, one would sometimes find MM and similar pedals set for the longest possible delay time, and often at the edge of audible whine (the lower the clock frequency, the longer the delay time).  I guess the assumption was that if the player didn't like the whine, they'd just turn the treble down at the amp.  So, it is also possible that the balance trimmer is just fine, but the unit has been set for longer delay than it can really manage.

That's a very early issue Memory Man, with only a "boost" switch to adjust input levels, rather than a variable input control as on later issues.  I'd be surprised if there was a compander chip in there.

You would do well to take and post a gutshot, so that we can recommend what trimmers to play with and how.

[Drewmeyer]

Yes.

The distortion, as described, sounds like the bias trimpot is off.  When the bias is set right, the wet signal is relatively clean.  When set a little too high or low, you can hear the delay but its distorted.  When set higher or lower than that, you don't hear any delay at all.

The audible whine sounds like it could be due to one or both of the following: need to adjust the clock frequency upwards a bit, need to tweak the balance trimmer on the output of the BBD so that maximum clock signal cancellation is achieved.  My own memory may be playing tricks on me, but if I recall the late 70's with any accuracy, one would sometimes find MM and similar pedals set for the longest possible delay time, and often at the edge of audible whine (the lower the clock frequency, the longer the delay time).  I guess the assumption was that if the player didn't like the whine, they'd just turn the treble down at the amp.  So, it is also possible that the balance trimmer is just fine, but the unit has been set for longer delay than it can really manage.

That's a very early issue Memory Man, with only a "boost" switch to adjust input levels, rather than a variable input control as on later issues.  I'd be surprised if there was a compander chip in there.

You would do well to take and post a gutshot, so that we can recommend what trimmers to play with and how.

Wow, thanks for all the info! Sounds like a calibration will do just the trick. Here's the gut shot:

Also, was EHX always this sloppy with their wiring?

[theehman]

pic not showing but yes, EH was always sloppy with their wiring.

[Mark Hammer]

Not to mention "generous" with their solder.  SPREAD THE LEAD!!    Actually, open up a larger MXR unit from that era and you would have seen the same sort of thing.  It was widespread.  As were the "flying components" on the backside of PCBs.

I'm seeing traces for 6 chips; a pretty sparse circuit, compared to later issues.  The chip on the far right of the picture is likely the BBD, with the trimmer on the "top" of the chip (in the picture) being the balance trimmer, and the one to the right and lower down the bias trimpot.

The pink stuff around the trimmers is the stuff that techs use to secure the trimpot to the position it was set to.  ONE of the trimmers - at least from the pic - appears to have considerably less of the stuff.  This suggests that it might have chipped off and that particular trimmer drifted from its original setting.  Knowing what chip/s it is near or connected to would provide a better sense of what it likely does and whether it might be responsible for the whining and/or distortion.

My own first step here would be to grab an old toothbrush, some methyl hydrate, and some cotton swabs or paper towels, and get all that flux residue off, so that I could see what I was doing.

[Scruffie]

wow, that's the REALLY old one with SAD1024s and no compander isn't it? Think You could post a component side gut shot?

If you search... I recently asked for a schematic of this and Stephen Giles provided me with one, there are 3 BBDs to Bias, if I find time (you may have to remind me) I can trace it a little to tell you which trim is which.

If you have access to a scope this'll be a lot easier... In fact I might not tackle it without.

There is no clock trim to get the whine out, it may come out slightly with the balance trim but with no companding and not masses of filtering, it's not surprising. You could lower the clock cap... but you may as well just turn the delay time down, it'd have the same effect.

Distortion might come out a bit with the gain adjustment too but feedback might suffer.

Edit: Okay, think I got it, Top trim near the red wires is BBD output Gain, Bottom Trim near the EH(Can't make out the number) is the Balance Trim for the 3rd BBD. In the middle left to right is Bias 1,2 & 3 and the one a bit up from those is BBD make up gain between BBDs 1 & 2

[Mark Hammer]

wow, that's the REALLY old one with SAD1024s and no compander isn't it? Think You could post a component side gut shot?

If you search... I recently asked for a schematic of this and Stephen Giles provided me with one, there are 3 BBDs to Bias, if I find time (you may have to remind me) I can trace it a little to tell you which trim is which.

If you have access to a scope this'll be a lot easier... In fact I might not tackle it without.

There is no clock trim to get the whine out, it may come out slightly with the balance trim but with no companding and not masses of filtering, it's not surprising. You could lower the clock cap... but you may as well just turn the delay time down, it'd have the same effect.

Distortion might come out a bit with the gain adjustment too but feedback might suffer.

Edit: Okay, think I got it, Top trim near the red wires is BBD output Gain, Bottom Trim near the EH(Can't make out the number) is the Balance Trim for the 3rd BBD. In the middle left to right is Bias 1,2 & 3 and the one a bit up from those is BBD make up gain between BBDs 1 & 2

Very good call. 

I don't have the schematic handy.  Safe to assume the 4th chip between the leftmost and middle BBD is a 4013?

Looking at the trimpot on the right by the EH.  It is tied to whatever is 4 pins from the end on that side of the chip.  Looking at the pinout of the SAD1024, this would be  "NC" (no connection) pins, counting back from one side, but are output A/B pins counting back from the end the other way.  So, "balance pot for BBD 3"?  Probably a safe bet.

[Scruffie]

wow, that's the REALLY old one with SAD1024s and no compander isn't it? Think You could post a component side gut shot?

If you search... I recently asked for a schematic of this and Stephen Giles provided me with one, there are 3 BBDs to Bias, if I find time (you may have to remind me) I can trace it a little to tell you which trim is which.

If you have access to a scope this'll be a lot easier... In fact I might not tackle it without.

There is no clock trim to get the whine out, it may come out slightly with the balance trim but with no companding and not masses of filtering, it's not surprising. You could lower the clock cap... but you may as well just turn the delay time down, it'd have the same effect.

Distortion might come out a bit with the gain adjustment too but feedback might suffer.

Edit: Okay, think I got it, Top trim near the red wires is BBD output Gain, Bottom Trim near the EH(Can't make out the number) is the Balance Trim for the 3rd BBD. In the middle left to right is Bias 1,2 & 3 and the one a bit up from those is BBD make up gain between BBDs 1 & 2

Very good call. 

I don't have the schematic handy.  Safe to assume the 4th chip between the leftmost and middle BBD is a 4013?

Looking at the trimpot on the right by the EH.  It is tied to whatever is 4 pins from the end on that side of the chip.  Looking at the pinout of the SAD1024, this would be  "NC" (no connection) pins, counting back from one side, but are output A/B pins counting back from the end the other way.  So, "balance pot for BBD 3"?  Probably a safe bet.

Not quite... 1st chip to the left is a 4047 (you can see the delay pot next to it tied to pins 2 & 3, one lug through a resistor) Then as you go right it's SAD 1,2 & 3. Balance is connected to Pins 5 & 6 of the SAD1024 so if we look at that far right chip, on the right side from top to bottom is pins 1,2,3etc... and it connects to 5 & 6! (little hard to spot the connection, but it's there).

[Drewmeyer]

Wow, you guys are awesome. Maybe a little too awesome. Some of that terminology I'm stilling trying to sort out. But, by request, here's a shot of the components.
Hopefully that helps someone out. I can post any additional pictures if they're needed. Also, is the general consensus that I should have a scope before trying to adjust the trims? Or should I try my hand at it and just go with what ends up sounding good?

[Scruffie]

The component side shot was actually for me, wanted to compare it to a schematic as i've never seen a component shot online of these old models does confirm what I said though... if you can get more shots it'd certainly be appreciated though, including under the pots... as would a demo as i've never heard one either but that might be asking a bit too much haha.

You COULD bias it without a scope if you built an audio probe, you could first (following strict ESD precautions and being very careful unless you wanted to buy new SAD1024s) remove the 2nd and 3rd BBD and probe after the 1st BBDs output cap with a some kind of input signal and adjust that trim for the least distortion, add in the 2nd BBD and do the same along with adjusting the make up gain trim and then the same again for the third followed by adjusting the balance trim for the least whine.

You can try and do it by ear but to get it biased 'right' will NOT be easy, the bias trims will only have a short section where they actually work and then need adjusting for the least distortion which when going through another 3 possibly misbiased BBDs is going to be hard.

I suggest either a scope or sending it to someone who has one.

[Mark Hammer]

Given the age (35+ years), there is a good chance that you'll want, or need, to replace the electrolytic caps, which may be a little dry.  The big fat ones may well be easily replaceable with something much smaller, with a higher voltage rating, for peanuts.

Biasing multiple-BBD units by ear can be done, but it is not easy.  The basic strategy is to to remove the ones not yet biased, somehow jumper from the output of the first to the point where the last one's output wouldnormally be, and treat the unit like it was a single-BBD circuit.  Once that first one is fine, you theninsert the second, re-jumper, bias, then insert the last one and bias that one.  The basic problem is that if you leave them all in place and tweaking one of the bias trimpots yields no improvement, you may have messed up that bias in the process of listening to the misbias of a different chip.  Confusing.  So, like I say, feasible to do it by ear, but not easy.  It may be far easier to do it by scope.

I have no schematic for this one among my holdings.  I've got later 3-Reticon issues, with companding, but none without it, as shown. Anybody have any pointers to one?

[Scruffie]

Given the age (35+ years), there is a good chance that you'll want, or need, to replace the electrolytic caps, which may be a little dry.  The big fat ones may well be easily replaceable with something much smaller, with a higher voltage rating, for peanuts.

Biasing multiple-BBD units by ear can be done, but it is not easy.  The basic strategy is to to remove the ones not yet biased, somehow jumper from the output of the first to the point where the last one's output wouldnormally be, and treat the unit like it was a single-BBD circuit.  Once that first one is fine, you theninsert the second, re-jumper, bias, then insert the last one and bias that one.  The basic problem is that if you leave them all in place and tweaking one of the bias trimpots yields no improvement, you may have messed up that bias in the process of listening to the misbias of a different chip.  Confusing.  So, like I say, feasible to do it by ear, but not easy.  It may be far easier to do it by scope.

I have no schematic for this one among my holdings.  I've got later 3-Reticon issues, with companding, but none without it, as shown. Anybody have any pointers to one?

Ah yes, that's another possibly better method than audio probing each chip but the same principal.

Yes there is a link to the schematic on this very forum provided by StephenGiles (actually for a different model # but practically the same) that I had asked for... where it is though...

[Drewmeyer]

Given the age (35+ years), there is a good chance that you'll want, or need, to replace the electrolytic caps, which may be a little dry.  The big fat ones may well be easily replaceable with something much smaller, with a higher voltage rating, for peanuts.

Biasing multiple-BBD units by ear can be done, but it is not easy.  The basic strategy is to to remove the ones not yet biased, somehow jumper from the output of the first to the point where the last one's output wouldnormally be, and treat the unit like it was a single-BBD circuit.  Once that first one is fine, you theninsert the second, re-jumper, bias, then insert the last one and bias that one.  The basic problem is that if you leave them all in place and tweaking one of the bias trimpots yields no improvement, you may have messed up that bias in the process of listening to the misbias of a different chip.  Confusing.  So, like I say, feasible to do it by ear, but not easy.  It may be far easier to do it by scope.

I have no schematic for this one among my holdings.  I've got later 3-Reticon issues, with companding, but none without it, as shown. Anybody have any pointers to one?

Other than increasing reliability, will changing the lytics affect the tone positively? And will it hurt the value? As I'm playing it a bit, it seems like distortion isn't as much of a problem as I had thought it was after an initial trial of the pedal. The main thing I'm noticing now is just the clock whine (I guess is the term, yes?). It's pretty bad once the delay knob is above half way. Is there any way to decrease/eliminate that?

[Scruffie]

Changing them may help the noise level (not sure on whine... might I guess) and I shouldn't think it'd affect the value too much, might even help it if you word it right, it's not too invasive such as drilling for an LED and the parts will eventually fail so people would probably have to get them replaced or have a dead unit anyway.

That whine (and probably the reason why it has a bit of distortion) is going to be down to the fact the pedal has no companding and very little filtering, that balance knob I mentioned will be really your best shot for getting some out. The chips just weren't designed to provide the 300mS this pedal is 'supposed' to provide hence why the whine creeps in as the delay time gets longer and the clock gets in to the audible range.

As a demo is most likely asking too much, could you give any description on sound (other than whiney  )? Very curious as i've never heard one.

[Mark Hammer]

When people recap old amps, they often save the caps that were removed, so that if any future purchaser has a hankering for that warm vintage 60hz tone, they can simply put the originals back in.

Improving the power supply caps will assist in keeping hum out of the audio path a little bit better.  There shouuld be nothing abut a contemporary 1uf electrolytic that is in any way better or worse, tone-wise, than a 35 year-old one, except that it is more likely to be performing to spec.

[Drewmeyer]

Changing them may help the noise level (not sure on whine... might I guess) and I shouldn't think it'd affect the value too much, might even help it if you word it right, it's not too invasive such as drilling for an LED and the parts will eventually fail so people would probably have to get them replaced or have a dead unit anyway.

That whine (and probably the reason why it has a bit of distortion) is going to be down to the fact the pedal has no companding and very little filtering, that balance knob I mentioned will be really your best shot for getting some out. The chips just weren't designed to provide the 300mS this pedal is 'supposed' to provide hence why the whine creeps in as the delay time gets longer and the clock gets in to the audible range.

As a demo is most likely asking too much, could you give any description on sound (other than whiney  )? Very curious as i've never heard one.

Well then I'll look into those caps. Also, honestly, I would love to record a demo since this seems like a fairly rare pedal, but I just don't have the equipment at the moment for recording anything of quality. If I get around to it, I'll be sure to send you the samples. But other than that, I think it has a really good sound. Repeats are shorter than I expected but are good for slapback and add a really good ambiance to rhythm parts if set to a good delay time. Can really nail that Edge dotted eight thing (if you know what I mean). The repeats sound beautiful. Really fat and chimey. The only thing I would want from this pedal is more delay time. But for it does, it does really well. I just need to find a way to get rid of the clock noise now. Also, what exactly does the boost function do? It just seems to make everything louder and cause the delays to clip and distort.

[Scruffie]

Best guess? It was put in for bringing up the signal of weak pickups or other instruments that didn't have enough signal (line in? This isn't my bag really, Mark will know more)... cause yeah it just boosts the input signal... for more level an output boost would be more useful.

It's not a long delay pedal sadly, this was literally one of the first analog delays to hit the market after BBDs were invented so it didn't offer what say... a Deluxe Memory Man or Boss DM-2 might offer, i'm betting guys with tape delays were pretty happy with 150mS of useable time in a foot pedal.

Thanks for the description! I had read that the Blue 4 knob Deluxe sounded 'ghostly' and it intrigued me... I was considering building one up with modern v3207 or 8 chips if it had a 'unique' sound from it's lack of filtering etc.

[Drewmeyer]

Best guess? It was put in for bringing up the signal of weak pickups or other instruments that didn't have enough signal (line in? This isn't my bag really, Mark will know more)... cause yeah it just boosts the input signal... for more level an output boost would be more useful.

It's not a long delay pedal sadly, this was literally one of the first analog delays to hit the market after BBDs were invented so it didn't offer what say... a Deluxe Memory Man or Boss DM-2 might offer, i'm betting guys with tape delays were pretty happy with 150mS of useable time in a foot pedal.

Thanks for the description! I had read that the Blue 4 knob Deluxe sounded 'ghostly' and it intrigued me... I was considering building one up with modern v3207 or 8 chips if it had a 'unique' sound from it's lack of filtering etc.

Yeah, interestingly enough, I, too, have heard Memory Mans be described as eerie or, as you're saying, ghostly, but really this one doesn't have that vibe, to me at least. It sounds clean, clear and quite beautiful. Also, that would make sense about the boost switch. I'm using it with humbuckers so I'm sure it doesn't need more input gain. But with single coils using the boost might be beneficial.

[Scruffie]

Best guess? It was put in for bringing up the signal of weak pickups or other instruments that didn't have enough signal (line in? This isn't my bag really, Mark will know more)... cause yeah it just boosts the input signal... for more level an output boost would be more useful.

It's not a long delay pedal sadly, this was literally one of the first analog delays to hit the market after BBDs were invented so it didn't offer what say... a Deluxe Memory Man or Boss DM-2 might offer, i'm betting guys with tape delays were pretty happy with 150mS of useable time in a foot pedal.

Thanks for the description! I had read that the Blue 4 knob Deluxe sounded 'ghostly' and it intrigued me... I was considering building one up with modern v3207 or 8 chips if it had a 'unique' sound from it's lack of filtering etc.

Yeah, interestingly enough, I, too, have heard Memory Mans be described as eerie or, as you're saying, ghostly, but really this one doesn't have that vibe, to me at least. It sounds clean, clear and quite beautiful. Also, that would make sense about the boost switch. I'm using it with humbuckers so I'm sure it doesn't need more input gain. But with single coils using the boost might be beneficial.

Funnily enough again, i've heard the 'update' of this unit, the Echo Chorus with Boost is described as a clean and clear delay too, that used a single MN3005 and a bit more filtering but still no compander (EHX switched to the Panasonic chips e.g MN300X instead of SAD1024s for both longer delays in a single chip and better noise) pretty sure they whine too just a bit less from added filtering, different designers though I think. The 4 knob memory man was supposedly 'ghostly' or eerie as you've read... eerie intrigued my style.

Single coils back then were probably even weaker so yeah no surprise, most BBD pedals can distort with modern high output humbuckers, especially active, I guess it was built in 'active'.

If I find time i'll try this with 3 x 3208, could half the clock cap and get the same delay without the noise, shame there's no such thing as SAD2048s you could drop in.

[Drewmeyer]

Best guess? It was put in for bringing up the signal of weak pickups or other instruments that didn't have enough signal (line in? This isn't my bag really, Mark will know more)... cause yeah it just boosts the input signal... for more level an output boost would be more useful.

It's not a long delay pedal sadly, this was literally one of the first analog delays to hit the market after BBDs were invented so it didn't offer what say... a Deluxe Memory Man or Boss DM-2 might offer, i'm betting guys with tape delays were pretty happy with 150mS of useable time in a foot pedal.

Thanks for the description! I had read that the Blue 4 knob Deluxe sounded 'ghostly' and it intrigued me... I was considering building one up with modern v3207 or 8 chips if it had a 'unique' sound from it's lack of filtering etc.

Yeah, interestingly enough, I, too, have heard Memory Mans be described as eerie or, as you're saying, ghostly, but really this one doesn't have that vibe, to me at least. It sounds clean, clear and quite beautiful. Also, that would make sense about the boost switch. I'm using it with humbuckers so I'm sure it doesn't need more input gain. But with single coils using the boost might be beneficial.

Funnily enough again, i've heard the 'update' of this unit, the Echo Chorus with Boost is described as a clean and clear delay too, that used a single MN3005 and a bit more filtering but still no compander (EHX switched to the Panasonic chips e.g MN300X instead of SAD1024s for both longer delays in a single chip and better noise) pretty sure they whine too just a bit less from added filtering, different designers though I think. The 4 knob memory man was supposedly 'ghostly' or eerie as you've read... eerie intrigued my style.

Single coils back then were probably even weaker so yeah no surprise, most BBD pedals can distort with modern high output humbuckers, especially active, I guess it was built in 'active'.

If I find time i'll try this with 3 x 3208, could half the clock cap and get the same delay without the noise, shame there's no such thing as SAD2048s you could drop in.

What do you mean by "drop in"?