different honeydripper problem

[pinkjimiphoton]

hey guys,
i built the "new " version of brian's pedal recently after (foolishly) selling the original one.
it just doesn't seem to quack right. i was like... wtf.
but "between" positions on the switch i can hear formanty goodness, plenty of yoy, but no real "quack"
so i thought maybe i put the rotary switch in backwards... not sure, i thought they were supposed to be marked, but there's only a "7" on the bottom of the switch.
so i desoldered it, rotated 180 degrees and now i got plenty of quack on the first two positions, and nothing on the other two.
planning on re-flowing the board with some fresh solder as i found a couple "microphonic" monolythic ceramics, so i assume the solder on them isn't great..
but do ya think i smoked the switch?
i DID buy two, so i can re-install a fresh one, but i don't wanna waste it when i can use it for another build.
any advice?
i will try and post voltages etc if anyone thinks they're necessary, but i'm banking on it being the switch.
thanks for the advice!
rock on and keep it quackin'!
PjP

[midwayfair]

What's the exact switch you used? Or was it from back when Brian was including them with the boards? Or are you using something you had lying around?

Do some continuity checks with the switch lugs. It's perfectly possible that the switch is boned, even though that would be pretty unusual.

Some pinktures and voltages/audio probe results will be needed for more anything more drastic, you know the drill.

[pinkjimiphoton]

yessir, i'll get to it tomorrow at some point. i didn't think it mattered which way the switch was mounted, but perhaps one way is break-before-make and the other is make-before-break or something.
it's the reccomended switch for the project from small bear, the one brian was including with the project originally.
i wanna build one of the older ones too and compare, so that will be fun.
this thing sounds great in the two positions where it works.
just got back from a 10 hour rehearsal with a rhythm section that was just not happening til the end so i'm toast now.
will try and get voltages, etc tomorrow.
gonna be hard to test the switch i think without removing it. we'll see.
thanks for the advice jon good to cross paths again man

[pinkjimiphoton]

ok, took it apart again, reflowed anything questionable including the "microphonic" caps.
no change.
replaced the switch, being sure to notice the numbers (on the top, under the washer) were right.

absolutely no change. thought for sure i'd found the problem... ic5c and d were way lower voltage than i expected..
until i looked at the schematic.

first two positions (yeh and bow) both kinda work. they sweep now, and sound slightly different.
3 and 4 (wah and aee) sound like a stuck wah.

i DID make a couple modifications to the circuit...

i used a 10k trimmer for r21 so i could dial in a sweet spot for the filter sensitivity... seems best around 1-2k, doesn't really do much anyway imho

i made the diode clipper half 1n34x2 in series and 1 1n914

c2 15n

d3 and d4 went again with the 1n34's thinking (maybe wrongly) it would make it more touch sensitive with a lower threshold. it seems to have worked, but...

ic voltages

9.6v wall wart

charge pump

1 9.50
2 4.51
3 0
4 0
5 0
6 4.54
7 5.97
8 9.50 (i shorted pins 1 and 8 together to nuke a cp whine)

13x00 (i've tried both, no diff, 13700 in there right now)

1 .44
2 .58
3 4.52
4 4.52
5 4.52
6 0
7 0
8 .47
9 .29
10 0
11 17.88
12 4.52
13 4.52
14 4.52
15 .40
16 1.30

1st lm324 (where the input is)

1 4.52
2 4.52
3 4.52
4 17.88
5 4.48
6 4.52
7 4.51
8 4.52
9 4.53
10 4.52
11 0
12 4.52
13 4.63
14 4.53

second lm324

1 4.51
2 4.51
3 4.52
4 17.88
5 4.32
6 4.53
7 4.53
8 2.07
9 2.50
10 2.50
11 0
12 2.50
13 2.50
14 .49

4558

1 4.52
2 4.52
3 4.50
4 0
5 4.50
6 4.52
7 4.27
8 17.88

to me, everything looks pretty good. perhaps i got the wrong switch somehow? both act identically...

stumped so any help is wicked appreciated!!!!!

thanks!

[pinkjimiphoton]

this was the switch i ordered, from small bear.
the original link in the build doc doesn't work anymore (of course)
but i believe it is the correct switch

http://www.smallbear-electronics.mybigcommerce.com/rotary-switch-miniature-2p4t/

[jimilee]

Kind of curious how this turns out, I was having the same issue, never did get it sorted.


Sent from my iPhone using Tapatalk

[bsoncini]

I built one a bit ago and mine was acting similar. It's waiting in a box for me to get back to it.

[madbean]

...not sure, i thought they were supposed to be marked, but there's only a "7" on the bottom of the switch.

Well this seems the most likely clue unless I am misreading this somehow. The markings should show 1-8 just like the picture you posted. The A and B poles are between the 2, 3 and 6,7 terminals resp (although on the PCB layout they are labeled 1-4 for each pole). These can't be installed backwards since the rotational connections are symmetrical.

Just to be sure I check continuity between poles and terminals on one of my switches and they are correct (plus I've built this version, of course). I shipped the switches with the first 100 boards I sold but after that smallbear started carrying them so I discontinued that.

Anyway, I would take one of the bare switches you have an do the same continuity test: in the left most position pole A should connect to terminal 1 and pole B to terminal 4. Then as you rotate through the positions they should connect with the next sequential terminals. If all else fails, remove the existing switch from the board and tack solder wires between the correct poles and terminals to check each position (although this would be really tedious).

[pinkjimiphoton]

thanks bri,
i'm out of switches, but i don't think the issue is the switch, unless maybe there's traces on the top of the board where the switch is hiding them from view.

i have audio everywhere, its quacking nicely in the first two positions, but the other two are a static filter.
i'm wondering if maybe the switches small bear has are slightly different?

i did find the numbers... they were obscured by the beauty washer that goes under the switch's nut.

i beep'd the original switch, and it seems to be functioning correctly.

my voltages seem good, other than the "b" side of the second lm324, its not even close to the 5v or so you
specc'd in the build doc... i'm reading less than half voltage there.

i' ve tried every quad oa i have, no difference. looked at the schematic til my eyeballs were bleeding, checked continuity, de-soldered stuff to measure, and it all seems to be right. so i am stumped!!

i boxed it up, as i only really used one of the positions anyways live usually, but its weird... the two positions that DO work both sound almost identical, and seem to be a combination of two different settings each. really peculiar.

i gotta gig tonite, so i'll try and get it apart over the next couple days again, and will explore the switching.

to me, it seems that the two sides of the switch may do what i'm experiencing if they are 180 degrees out of phase with each other somehow... that would put all 4 possible connections from two on one side and two on the other to all 4 being on at once in the first two positions. i THINK.... if looking at the board and circuit is right.

since its identical with both switches, i wonder if its the switch that's the issue? i've found often that when purchasing chinese components - and face it, thats where everything seems to come from- often there are mistakes made.
i'm wondering if maybe they screwed up the internals of the switch?

i don't see how it can be the board or the circuit if its passing signal and the voltages are within a reasonable tolerance.

perhaps i should seek out an actual alpha switch instead of the small bear knock off?

weird. curiouser and curiouser!!

thanks for the replies guys... there's a few of us with this issue now so it doubles my resolve to conquer it!!!

SQUASH THAT DAMN BUG!!!

[bsoncini]

I didn't buy my switch from small bear. To expensive to ship to Europe.  Got a 20 pack for cheap on eBay from china.  I have used at least 7 of them and have always tested them before using because they look really cheap. Definitely tested before the honey dripper and never had a problem with one.

[madbean]

my voltages seem good, other than the "b" side of the second lm324, its not even close to the 5v or so you
specc'd in the build doc... i'm reading less than half voltage there.

B-side voltages on that LM324 will change some through the different switch positions which I totally failed to mention in the build doc. So, it's possible I recorded my voltages with the switch in a different position. And, there may be some other pins changing voltages through the different positions as well. Do you remember what setting you were on when you recorded yours?

All things being equal, assuming all parts and values are correct it's hard to say exactly where the problem is. My inclination is to blame the switch or some kind of board-level failure (it's rare but it can happen). If it were me, my next step I would take is to remove the switch and hard-wire either positions 3 or 4 of the rotary to rule out the switch itself. If you can remove the switch without damage then it won't be too hard - jumper SW1A pole to both pins 3&4 A terminals and then jumper SW1B pole to either B terminal 3 or 4 to test the third and fourth positions, resp.

If you can get it working that way to at least confirm the switch as the issue I can send you one of my spare rotaries (which I know are good).

Lastly, from you voltages post: are you using the LT1054? You should not have had to jumper pins 1 & 8 with the LT1054 (that's only required when using other types of charge pumps to enable the internal frequency boost).

Sorry this has been such a bummer. I've re-checked over my schematic, board layout, parts library, switch functionality and my own personal build just to be "quadrupley" sure there's not some design flaw and I can't find any.

[pinkjimiphoton]

thanks bri,
i am absolutely sure i f'd something up. i am more than that sure that the project isn't at fault.
i had a gig last nite and just got home...had to work all day so i'm too bushed to check tonite,
but will try and poke around in it on sunday when i get a chance.
the voltage is not changing on the quad with switch changes, so i'm betting it's either the switch, or i damaged a trace when i replaced it.
thats why i was asking if there were traces on both sides of the board.
i have no doubt we'll sus it out.
thanks for the support and kindness, they're very appreciated and will be paid forward.
i'll post as soon as i can.
peace!

[pinkjimiphoton]

sorry, life's been happening.
beginning to think the 13700 is bad. i got some from tayda, and had recently gotta swah from barry's site and it uses the same chip. and that's not sweeping either.
gonna try to get to some bench time soon and sort this out. looking at it, all the voltages seem right, so wondering if i wasted a trace somewhere i can't seem to find. i blew up the graphic pretty big and seem to have continuity everywhere as expected, so this is really weird.
i've tried a bunch of different quad oa's same dif, no real difference.
but the fact the other build is also failing is making me lean toward crummy chips.
when i get a minute i will try and order a couple from a reputable vendor <for a change>
and see if that makes a diff. both chips are from the same vendor, and same order.. i usually try and buy everything 10 at a time at least when ordering components.
thanks guys. it will be soon i hope. peace!
ps. brian, thanks for the boards for the old style HD, i hope to get one populated soon!

[pinkjimiphoton]

thanks for the patience as i was camping on this one. rough period of time culminating in my brother, best friend and bandleader Eddie movin on across the river styx.
anyways
i finally gotta chance to take a crack at it tonite.
i tried a shit tonne of opamps in it, everything i could think of that was on hand.
no change. started suspecting the switch when i found a couple bad solder joints.
turned out the switch was fine, but the traces were wasted from being reworked so much.
i tend to be REAL fast and use a 30 watt iron for most things, which is foolish i guess but i'm usually pretty
good at not melting shit too bad. sometimes. mostly, these days. when i was starting.. you know the drill.

anyways, i took the switch out again. its actually fine, believe it or not. but i found the problems.
i take responsibility for them, tho i suspect that if others were having the same exact issue, it may have been a prob with the manufactured board. but i think it was likely me.
the common a and b pole sockets were wasted. i thought they were plated thru hole, but in this it appears it wasn't. doesn't matter, i had lifted the pads when i replaced the switch the first two times, and since it was under the switch was too dumb to notice.
once the switch was out, i just did simple continuity point to point checks... if you guys having problems look close, brian has test points freekin everywhere that help breeze-ify the debugging process, a really nice touch.
in addition to the a and b common poles being cheezed... which is why i only had one sound really, but in the first positions.... a couple of the other pins pads were damaged too, which led to loss of continuity.
so i put the new switch in brian sent me... dude, yer a gem... and made sure they made contact with R26... before they DIDN't. so it couldn't quack.
the trace between pads 2 and 4 of switch b had been damaged, as had the trace to  the right side of R25.
so i added a jumper between pads 2 and 4, and tack soldered that to the node on R25, and yawza wowza  we gots us some formanty goodness. sounds pretty good, too. i think the original reacts a bit differently to your attack, but that could be from me using germanium. the sensitivity control needs to be a bit more cranked than on the other one, but its' sounding great now.
take the time and look at the board layout in the build doc. follow voltages, anywhere suspect figure out what it is that can be preventing it from being right. eliminate one thing at a time, and eventually you will find the issue.
funny that the switches are all still good for future projects.
i hope this info helps someone else, as i imagine if the same issue is happening, this is why.
it looks like on only a couple of the holes, the thru-plating between the solder pads was missing, which may be a manufacturing defect from the place that makes the boards.
like i said, not pointing fingers as i am fairly sure i fucked this up and i wanna own it, but this may suggest why a few other folks had the same issue. its easy to mess up when ya got two sides lol... ya need some penetration!!
but anyways, awesome project as was the last, thank you all for the support!!
and especially brian, the madbean himself.
rock the fuck on peeps, and happy holidays no matter what or how you roll
gotta hit the rack, doing a big benefit for your friend and mine, buddy cage from new riders of the purple sage. 5 bands, i'm donating a guitar, all kindsa insanity. 12 hours away. i gotta get some sleep.
peace out!

[midwayfair]

Sorry to hear about your bandmate, Jimi.