The 'Gilmour' Build Guide (Deluxe Electric Mistress V4 Clone)

[stallik]

The board is 80% populated but it seems every time I get chance to work on it I need something else. Popped into Maplin today to collect a couple of wrong caps for the second time!!
That's a bit of an excuse really- my wife got me a Ditto looper as a birthday prezzie and its really easy to play instead of build.....

[stallik]

It's coming on - missing the 1N5817 diode and a couple of IC's that have been on order for ages...
Also giving up on the 15uf bipolar as the only ones I've found are pretty big. Going for the pair of 33uf's





Used flexible wire for the pots so that I can flip the PCB up and am going to put a stand off at the top though it all seems pretty robust as it is.

[Govmnt_Lacky]

Anything new Kevin? 

[stallik]

Yes. Apparently, the seeds have been planted to grow the trees, to turn into carbon, to use for the chips I stupidly ordered from the other side of the galaxy...
The longer I wait, the less patience I have but if I give up and order some from the local supplier, they'll cost more and the first ones are bound to turn up as soon as I do.
When they arrive, I have to calibrate the pedal. I note that an oscilloscope is recommended but a MM + ear can be used. Was considering one of those tiny little ARM scopes on ebay. Maybe this is an excuse to get one. Anyone have any experience of them?

[stallik]

Chips have finally arrived, powered on, little effect, lots of debugging to do. Set my power supply voltage to give me 12.1v after the protection diode and measured voltages of ic pins. Most are correct, some are way out. Can't adjust pin 1 of u6 to 40khz, available range is 150 to 300. At some settings I'm getting an effect (just) put there's lots of popping etc. Suspect I've made a mistake somewhere
Also think that I've wired the pots the wrong way round as, for example, turning the rate pot counter clockwise increases the rate
Try to get to it tonight

[Govmnt_Lacky]

Chips have finally arrived, powered on, little effect, lots of debugging to do. Set my power supply voltage to give me 12.1v after the protection diode and measured voltages of ic pins. Most are correct, some are way out. Can't adjust pin 1 of u6 to 40khz, available range is 150 to 300. At some settings I'm getting an effect (just) put there's lots of popping etc. Suspect I've made a mistake somewhere
Also think that I've wired the pots the wrong way round as, for example, turning the rate pot counter clockwise increases the rate
Try to get to it tonight

Definitely should have a lot more range on the clock. Sounds like you do have the pot(s) wired backwards.

I believe there are some good voltages in the build document. Those are spot on for the working unit that I built.

Keep us informed of your progress on the debug 

[stallik]

Think I need some help.

Last nights debugging highlighted the following:

I had wired the pots the wrong way round - corrected
I had also wired the switch incorrectly - corrected
R3 is listed as 1M but all the layout images show colour bars indicating 100k. Changed it to 100k but it did not appear to change anything.

I have an effect sound. In Matrix mode, colour, rate and range appear to be working correctly. I have a wide sweep with the rate changing from a fast wobble to a wonderful slow (just how can it keep going?)
In filter mode, the range pot alters the tone & the colour pot can make a difference depending on how the trimmers are adjusted. So what's the issue?

1. The sound level when using the effect is very low - about half that of the bypassed sound and the effect sound is also somewhat muffled. It sounds almost like there is a cap to ground somewhere which is bleeding off the high frequencies

2. I cannot get down to 40KHz on pin 1 of U6 - I appear to be stuck with a range of about 150 - 300 when adjusting the trimmer but setting the range pot to full, I watch a sweep from 60 - 290.

3. I have some hissing - but the board is hanging out of the enclosure

4. I have some wayward voltages (highlighted in red)




My power supply is a 1.5A adjusted to give me 12v after the series reverse protection diode D9A
I've checked for shorts on the board using a powerful binocular scope and can't find any
Any advice gratefully recieved

[GGBB]

R3 is listed as 1M but all the layout images show colour bars indicating 100k. Changed it to 100k but it did not appear to change anything.

R3 is 1M and that is what is shown in the layouts - brown-black-black-yellow - but DIYLC's yellow is not much different than its orange unfortunately so R3 looks almost identical to R2 right below it which is 100K brown-black-black-orange.  I believe it just sets the input impedance so using 100K wouldn't change anything else circuit wise.

Sorry, I have no advice or help to offer other than the obvious checking of diode and polarized cap orientation.  Also, if you used 2 x 33μF polarized caps in series for C16 make sure the anodes are outside and the cathodes are tied together.  And C11 polarity is wrong in the schematic, but correct in the build guide.  Hopefully Greg will have some other suggestions.

[Govmnt_Lacky]

There still appears to be a problem with your clock. Most of the voltage differences look to reflect that.

Check to make sure that you have the 3K9 resistor (on the clock trim wiper) at the proper value. Also... double check that you are using a 100k linear POT for the Range.

Also, as Gord suggested... check to make sure your LFO caps are set correctly (if using dual 33uF caps... ensure you have the +'s tied together)

Your ambient noises are most likely due to it being out of the enclosure but, there will be a small amount of background noise as with most EHX effects. Nothing that cannot be squashed with a gate though 

Could you post a pic of the fully populated board?

[stallik]

Thanks GGBB, you are of course right about R3. Taken another look at the pdf and the yellow/orange are different. By the time I'd printed it out however... I'll put the 1M back over the weekend. Taken a look at the pic of my board (at work now) and believe C11 is correctly orientated, same for the back to back caps for C16

Gvmnt Lackey, will check all those points and get a pic of the board up late tonight

Hope to continue the debug over the weekend

thanks again

[GGBB]

Also, as Gord suggested... check to make sure your LFO caps are set correctly (if using dual 33uF caps... ensure you have the +'s tied together)

I am a little confused.  My understanding has been that the negatives need to be tied together.  That makes sense to me when I think about how the anode/positive side needs to be at a higher DC voltage than the cathode/negative side.  A google search seems to support me, but not exclusively - it is the internet after all.  Perhaps either way works?

[Govmnt_Lacky]

Just had another look at my Echoflanger build. It is the only LFO based effect that I have on-hand that uses a "back-to-back" electro cap design in the LFO. That design uses (2) 50uF caps with the positive leads tied together.

Also, if you look at the original schematic for the DEM V4, it shows the 15uF LFO cap (non-polar in the original) with the negative on each side [ ) ( ]

Of course... it is not a huge hassle to try it either way  

EDIT: @ stallik

Make sure you follow the build doc when adjusting the clock. I believe that the range pot needs to be either full CW or CCW AND the toggle needs to be in MATRIX when adjusting. Also, rewiring the pots might call for another adjustment of the clock trimmer  

[armdnrdy]

I have built working flangers that depicted the timing caps in the schematic both ways....cathode to cathode and anode to anode.
I think that either way turns the two caps into a non polar/bipolar.

[stallik]

OK guys, getting somewhere I think.
Over the weekend I've double checked for shorts, reflowed all the joins, rechecked for shorts, replaced the switch, replaced U2 with an alternative (can't remember which ic, let you know later) result has been the same. Nice sound, everything working but quiet effect sound. Then I applied an idiot filter and things came a little clearer.

1. I realized that I'd made a mistake on labeling the switch. Instead of Matrix/Flange I'd labeled it as Matrix/Filter. It appeared logical to me that the Filter setting was giving me the non LFO tone change while the matrix was the flange. This would not matter except I'd been trying to get 40mhz on the wrong setting. Switched it correctly and viola! 40mhz. BUT... This did not give me an increase in level.

2. I found a problem with my power supply cable - an intermittent, partial short which was really overheating the ps and causing it to go into emergency shutdown. sorted that BUT it did not give me an increase in level.

3. Tried replacing R17 with a link. Yippee! Increase in level. Checked the value of the resistor and it was to spec, probably something else causing the issue and leaving this out was only changing the symptoms, not the cause but decided to have a play. Late at night so only low volume but wow! Sounds really good with both clean and distorted tones. Removing the resistor has given me a lot of hiss in Matrix mode so I think this is only a temporary workround

This is the board before the modifications

[Govmnt_Lacky]

3. Tried replacing R17 with a link. Yippee! Increase in level. Checked the value of the resistor and it was to spec, probably something else causing the issue and leaving this out was only changing the symptoms, not the cause but decided to have a play. Late at night so only low volume but wow! Sounds really good with both clean and distorted tones. Removing the resistor has given me a lot of hiss in Matrix mode so I think this is only a temporary workround

A couple of suggestions...

- Try upping the value of C2 to 470uF. I found that using a 220uF did not filter out some of the "noise" as well. Once I installed a 470uF... it seemed to do the job.
- Try playing with the R17/R18 values and adjusting the Feedback and Mix trimmers to get the best out of it.
- Maybe try a different dual op-amp for U2 (TL072, NE5532, etc.

Good Luck and... let us know how it turns out for you!!  

[stallik]

THANKS GUYS!
Finally had chance to complete my build. Found that the volume loss was almost entirely down to R17. Ended up settling on 560r and 94nf for C20 (There's another 47nf on the other side of the board)
U2 is an NE5532P.
Clock is set to 40Mhz, Bias tuned by ear - wound it (down) till the signal distorted, then bought it back up till the distortion went, then added a bit for luck. Scientific huh?
Played around with Mix and found I could go from nothing through flange to almost synth. As this changed, the colour control went from no effect to quite a bit so settled on something inbetween where, with the colour down, good flange, colour up, synthy(ish). While playing with these 2, I kept adjusting feedback to stop oscillation with colour all the way up.
Found that with the setting I'd chosen, the mix can be set to a point where oscillation is just trying to happen and you can actually control this with pick attack on the guitar. Might pop in another switch....

I've left the switch labeled incorrectly. Matrix is flange, Filter is Matrix

So what does it sound like? Well you know those times when you get a sound that just seems to make you play better? I just love this pedal!!

Clean sound, Colour 40, Range30, Speed 50 or so - nice swirly, chorusy/ flange works well with all pickups on my Strat & Tele
Dirty: Spitfire-(PT80 delay)-Gilmour with Colour 100, Range 100, Speed 0 - Really addictive, the range of the sweep is incredible. 20mins in and the riffs are just inventing themselves

OK. Time to get sensible. I play crap. It's unlikely that any other guitarist would be impressed with what I've just been playing. I usually cannot play what I want to so just play what I can and have to settle for that. But I've just spent the last hour or so playing things that I cannot normally play. I know its only psycological but that, in my book marks this pedal as a definite keeper.

I'll try to get a soundclip sorted, in the meantime here's the final pics (the knobs are on - I only do that when everything else is done)



And I've left R17 & C20 on the trace side


Oh, by the way, this is the 12v Dc versionso I'd consider it verified

[Govmnt_Lacky]

I just LOVE that build. Perfect results!

Enjoy it bro! Its a killer pedal.

Cheers  

EDIT: Would you mind sharing your powering scheme? This way... others can see how to power it successfully WITHOUT the transformer. I see you labeled the pedal 18V. Are you regulating down to 12V inside or outside the pedal? Details my good man! 

[stallik]

Regulated down to 12v outside of the pedal. Originally was going to go with  with regulation inside but I have a 12v  supply to my board so decided to use that.
I went with the series diode reverse polarity protection and adjusted the external PS regulator to give me 12v after the diode. Interestingly, the pedal actually functions, though not as well, at 9v so adjusting for the voltage drop by the diode might not have been necessary.
The label is now wrong, hope it doesn't confuse anyone

[Govmnt_Lacky]

Regulated down to 12v outside of the pedal. Originally was going to go with  with regulation inside but I have a 12v  supply to my board so decided to use that.
I went with the series diode reverse polarity protection and adjusted the external PS regulator to give me 12v after the diode. Interestingly, the pedal actually functions, though not as well, at 9v so adjusting for the voltage drop by the diode might not have been necessary.
The label is now wrong, hope it doesn't confuse anyone

Thanks Kevin! 

What type of PS are you using? Just for the full story... 

I appreciate the feedback. I really don't know why more people are not building this  It is relatively cheap when compared to other flangers and it sounds AWESOME!! 

[stallik]

I cheat with my power supplies. Many years ago, I tried a Velleman 1.5a regulator board kit which I picked up at Maplin in the UK. It's output voltage is easily adjustable via a trimmer. I've found it to be very quiet even when daisy chaining and my first one is still working fine 15 yrs on!  They're still available and cost less than the components and pcb so I've never bothered to make my own
I have 2 supplies in service at the moment, one with a 2x12v transformer, 2 regulator boards, the other a 2x15v toroidal with only 1 board