OK, here is an image file with all the pin voltages I recorded on my 9V Electric Mistress build so far.
(http://i55.photobucket.com/albums/g154/sta63bmx/effects/9v_electric_mistress/9V_ELECTRIC_MISTRESS_VOLTAGES.jpg)
I do NOT have this boxed yet. It's laying out on my bench with a bunch of wires splayed everywhere. :) I've got a couple comments/questions, though.
Comments
1. I noticed that after I was done ironing, the pads were touching in some places that they shouldn't have been. I scraped the transfer off before I etched it and then it was ok. Smaller pad sizes might have helped with that? I dunno. The fine details came out in the iron and etch, though. I mean, the REALLY fine lines came out clean. Pretty cool! The ironing surface wasn't perfectly flat, so maybe I just got too much pressure in some areas.
2. Did it say in the project file which lugs were 1, 2 and 3? lol I must have missed it. I had all my pots wired backwards the first time. :D What's the convention when looking at the BACK of the pot? Is there a convention?
3. Tons of room for everyone on the board. No wacky squeezing or bending or anything.
Questions
1. What *exactly* is the clock trim setting? I read through that old copy of Device from Mark Hammer's page and looked through some search results and I am just not quite getting it. I apologize for my dumbness.
2. What the heck is up with that 2k7 resistor on the battery input pad? lol Is the battery supposed to be charging from the power supply when the unit is plugged in or...?
Sounds/Trimpots
1. The bias trim I found to be THE most critical. If it's out one way or the other, I got zero flanging. There was an area where I didn't hear any distortion, but it would clip at either end of that range. I couldn't hear any level change inside that sweet spot, and it was pretty wide. I didn't notice much change in the flanging sound.
2. The feedback trim controlled the depth of flanging and gave weird spaceship sounds when the unit was in oscillation. When the unit was close to oscillation, I noticed more white noise kind of sound. Is that from the clocking signal?
3. The clock trim was the most confusing to me. In the middle of the range was where I got the worst noise. I was running into a semi-cranked tube amp, there are long unshielded wires, I was expecting a little noise. No big deal. I am just not fully understanding what that knob does. I fooled around with the three pots for a while, adjusting the clock trim trimpot and I found my favorite sounds towards one end or the other. In the middle, the sounds I got were more "detuned" and less useful, with more wobble and weirdness. I was looking for that slower, more musical sound, and was better able to get that at one end of the trimpot's travel. Less clock noise, too.
4. The unit was very quiet where I had it set. Aftermy Pulsar Experience that was a pleasant surprise.
Overall Impressions
1. The non-true-bypass dry signal was VERY similar in gain to the flanged signal. They were close enough that I'd rather do non-true bypass switching. The tone sounded fine to me.
2. I really liked the slow, subtle sweep. The intro to "Mayonaise" by Smashing Pumpkins is in there, but so is "Love Stinks" by...jeez, Nazareth? But the slower sounds I liked best. It doesn't completely overpower the regular sound and you can still PLAY with it, not just make weird noises.
3. The vibrato was also handy.
Summary
It worked on the first try, so props to Markus for the layout and Fransisco and Steven (?) for the schematics. I'm sorry if I left anybody out. Now it's time to make a cool box. Where could I wire in an LED that would flash in time with the LFO? I plan to make a super-pimp status light scheme for this box. It's for my coousin's birthday and he has no idea. I'm super excited about giving him this present. I don't think I've ever been so excited about something I got for somebody. I'm honestly surprised I got it to work right away. Thanks for the layout and schematics, guys. This is the coolest present ever. Kinda. Oh, here's the board.
(http://i55.photobucket.com/albums/g154/sta63bmx/effects/9v_electric_mistress/9V_ELECTRIC_MISTRESS_SMALL.jpg)
Those readings were all made with a multimeter looking at DC voltage, by the way. I'm sorry I forgot to mention that.
Pin 7 of the LM324 provides a 50% duty cycle square wave at the same frequency as the LFO, so I think I can use that to switch the "flashing light" indicator that I want. Pin 7 is the output of one of the four opamps on that IC. This is where my glaring lack of electrical knowledge comes in. Those two sections of it with the 1M cap are the LFO, right? And is it just a square wave LFO with one opamp being driven back and forth to its rails?
I am ASS-SUMING that the output where I see that square wave can supply enough current to drive the base of a transistor without messing up the rest of the regular EM circuit. Is that true? I was going to connect an LED inline with the collector of some little NPN (got some 3904's handy) and then ground that emitter to the same ground as the on/off LED (so it would be switched by the footswitch) and THEN drive the base of that NPN with the output of pin 7 from the LM324. Do I need a base resistor or something? If someone could let me know what rules of thumb to follow here, I would appreciate it. The voltages already seem perfect for driving that gate on and off.
The LFO can go so slowly that I think a permanent "status" light is necessary, but the flashing light would be nice, too.
thanks so much for doing these and as well for even posting them here for all so that they WILL be here from here on. Since I know that you basically went through the trouble as favor to me thought I'd make it known it's greatly appreciated , you put some effort into it not just scribbling reading down! :icon_eek: I'm sure others will appreciate this file in thefuture
If this doesn't help me track my error/problem I don't know what will! and i'll update with my progress
the kid :icon_wink:
Dude, I owe everybody on the site so much for the free layouts and help that I could never pay it back. I can't wait until yours is up!
Thank You, David %^&*erell and Howard Davis
for having provided Mike Matthews with such fine pedals!
(like you did, and partly still do...)
:icon_wink:
I am striking out on getting a flashing light that I like. Pin 7 on the 324 is a 50% duty cycle square wave, which is perfect, but I CANNOT for the life of me get it to drive a transistor at 50% duty cycle. It goes from about 0.6V to about 5.4V and the frequency is controlled by the rate pot. Here's what I tried first.
1. Put a current limiting resistor and LED between V+ and the collector of a 3904
2. Hooked pin 7 of the 324 directly to the base of the 3904
3. Connected the emitter to ground
At best I got a very faint light flash, using a pullup or a pulldown. By fooling with the current limiting resistor I could get a brighter flash. But it was like a 1-10% duty cycle, tops, like a very brief flash. It was lame. So I looked at the signal on pin 7 of the 324 and tried to see what impedance it was.
I put a decade resistor between pin 7 of the 324 and ground, and I had to drop the resistance to less than 10k to see any change in the waveform. That signal should have plenty of oomph to drive the base of a transistor. So why won't it work? I'm trying to rig a transistor like the blinking light mod outlined here in the runoffgroove EA trem schematic: http://www.runoffgroove.com/eatremolo.html (http://www.runoffgroove.com/eatremolo.html)
What act of gross stupidity am I committing? I thought if the base was at least ~ 2V more positive than the emitter than the transistor would turn "on" and switch on the LED. If I monitor pin 7 on the scope while I connect it to the base of the transistor, I'm not seeing any change in the waveform. So it isn't like the transistor base is pulling that down to nothing. There must be something I'm doing wrong. Probably something dumb, knowing me...
I guess I'll ask for some opinion from the forum here as well and the wealth of knowledge here.
I have had problems getting mine to flange , originally getting only clean signal abslolutely no affect from any controls at all. but now I'm not getting any signal , so not sure what happened there.
Based on voltages though, All is well based on the voltages here, EXCEPT for the two ICs' SAD1024 and flip flop 4013. Of which the biggest mystery IS...
What possible causes could allow Pins 3 and 14 of the SAD 1024 and 2 and 5 of 4013 (which are only connected to one another) be reading at ZERO volts if not due to being shorted to ground anywhere?
My entire problem so far as I can tell seems to be coming from this area.
And I'll note , since the SAD 's are so rare , that I borrowed mine from a Ross Flanger I have and it does flange in that circuit as well as Pins 3 and 14 giving voltage of apprx 1/2v+ . So hope that tells me it's not the chip.
Any ideas??
Quick note, I tried using a 74128 NOR gate to drive the transistor base. Using the LM324 pin 7 as the A and B input to the NOR gate and the Y output to drive the base, I could make it work. The trouble is that I think the chip draws WAY too much current. I used a voltage divider (drawing off some power from the board like I would have done in the pedal) and I kept getting a HUGE voltage drop to the middle of the divider. I had to use a voltage divider to get 5V for Vcc on the chip, and I noticed that the voltage was *way* low. I think the chip is drawing something like 10-50 mA, probably more current than the whole board, maybe? It was nuts. I had a 133 ohm/5k divider, and right after that 133 ohm resistor, the voltage was down to about 5V from 7.2V.
I think I'm going to give up on the flashing light unless someone has a great idea. I'm just pissed because there is a perfect control signal RIGHT THERE and I can't seem to make use of it. Here's a picture of the signal on pin 7 of the LM324.
http://i55.photobucket.com/albums/g154/sta63bmx/effects/9v_electric_mistress/LM324_pin_7_small.jpg
I am just not sure how to implement that to drive a transistor. I thought I did, but evidently I don't!
I am the dumbest dumbass to ever be dumb. I think I am the king of "I can't get it to work!" and then ten minutes later "Fark, I'm dumb."
I was trying LAST night to hook that signal directly to the base of the transistor. Bear with me. This is for an NPN transistor.
1. The base-emitter junction just kind of looks like a diode. So the base will be about 0.7V above the emitter. And if you send a tiny amount of current into the base (does it actually go to the emitter? and doesn't Vbe have to be bigger than some threshold?) you can make a HUGE current go from the collector to the emitter.
2. By hooking the signal DIRECTLY to the base, I was forcing that pin to be no more than 0.7V above the emitter, which was tied directly to ground. So goodbye LFO signal. And then it would never have enough voltage from base to emitter to turn the thing ON, so no flashing light.
So today I hooked the transistor up EXACTLY like last night only I put a 360k resistor in series with the base. So now I can still get a tiny current but all the voltage drop occurs across that resistor. And so the LM324 pin keeps doing what it's doing, but now the light is flashing.
I fell asleep reading AofE last night, but I'm going back to the transistor chapter to try and figure out where I'm not quite GETTING it. I have always thought of "turning on" an NPN transistor by applying a positive voltage to the base relative to the emitter. The single diode drop from base to emitter would have killed the voltage, so no triggering. That makes sense. But is is the base current turning the thing on or is it the voltage relative to the emitter that turns it on?
I'm going to put the flashing light stuff and regular indicator light stuff on a little piece of perf. I am going to filter the PS for the flashing light some to try and make sure I don't pick up a pop.
Does that 200 ohm resistor between PSGND and V- float the ground plane relative to the negative battery terminal? This is how my voltages all measured.
Negative Battery Terminal: -1.8V
PS GND: 0V
PS +: 7.2V
Is that done to try and cut down on noise like how you float an AC filament supply on a DC bias? I'm not understanding the why of that particular resistor.
..... no ideas, anyone, on my voltage probs around the SAD and flip flop?? anyone experienced with these 1024stage BBD's particularly and what reasons besides short to ground could cause no vltge reading on pins 3 and 14.
Hello,
I'm trying to make the same circuit (electric mistress 9V - Markus W.).
But i have just clean tone when testing.
I've made some measures to compare with the ones here.
I saw a strange thing:
Negative Battery Terminal: - 4,6V
PS GND: 0V
PS +: 4,4V
I was suspecting some short on the board, but i've already verified everything.
I tested on two diferent boards with the same results.
What i notice is that when i take out the LM324, i get the correct values.
But i don't thing the CI is burned because i tried with diferent ones (maybe all of them are burned?).
When i take out the RC4558 i get almost the correct values too.
I've made some substitutions:
Resistors
200 R -> 220 R
510 R -> 560 R
6k2 -> 6k8
13k -> 12k
24k -> 22k
30k -> 33k
62k -> 68k
200k -> 220k
1M2 -> 1M5
Cap
1n -> 1n5
Diode
1N961B -> 1N4742A
I couldn't find the correct values, so i put what was the nearest.
I don't think this is the problem.
So if anybody could help me, i'd be glad.
Thanks a lot in advance,
Wagner.
In the original Mistress the 2k7 at battery V+ is bypassed if no ps is connected.
So easiest way is to connect both, battery and ps V+ to the PS+ pad.
Hope this helps.
hello Markus,
Thanks for the answer.
i'm not using battery... so i'm only connecting the PS...
The problem is with the 200 ohm resistor conectec to gound...
So my reading are:
ps - terminal = -4.6V
ps + terminal = 4.4 V
gnd = 0V
for some reason i think the circuit is eating a lot of current...
that's why the ps - is -4.6V... it should be around -1.8V...
i was thinking that is was a short or something like that...
but i have the same problem with 2 boards...
does anybody have any advice?
thanks,
Wagner.
BTW, I replaced the 200R with a jumper.
Regards,
Markus
Hello Markus.
Thanks for the answer.
I tried to replace with a jumper too.
This way i get higher voltage from the table, but proportional values.
But i get only clean tone not flanger.
Any more ideas i could check?
Thanks again,
Wagner.
Do you have a possibility to check if the LFO and the VCO are working?
Regards,
Markus
Hello Markus W.,
Thanks for replying.
Can you tell me how to do it?
I think i need to understand a little bit each part of the circuit because i don't know all parts yet.
Because i can get clean tone i can assume that the TL072 (input buffer) is working properly?
I know that the LM324 is working too because i can see that the values are changing according to the value tables above.
But in diferent values... and also i could check the square wave on pin 7 of this IC.
I have only a DMM... i don't have an osciloscope... but i can get one if it is needed.
What else could i check?
Thanks,
Wagner.
Hello,
I'm back again trying to make the electric mistress work.
I have now almost everything working great.
The effect is working very good.
Now the problem is with noise.
I'm getting what i think is a clock noise (noise in the same frequency as the oscilator).
I've used the biggest filtering caps as suggested on the PDF project file.
I also changed the CLOCK TRIM to get the lowest clock noise.
But I'm still getting CLOCK noise.
There is one point when the FEEDBACK and the RATE controls are on max (fully CW) and then when i change the RANGE control there is one point when the CLOCK noise goes away. That's teh way i think it should be for all setting.
Is there anything else i could do?
Thanks in advance,
Wagner.
Hey,
glad you got it working :)
Sorry I didn't get back to you. :icon_redface:
I left for holidays on July 11th and when I came back I simply forgot to check this thread...
Where was the bug finally?
Regarding your noise issue: does the frequency of the noise change at the rate of the LFO?
Do you happen to use a switchmode power supply, btw?
Peace,
Markus
Hello Markus,
Thanks for answering.
There was a lot of bugs... first my board, i don't know if i let it less time than needed on the acid, but i found some short circuit... but it was not my soldering... i found mainly on the IC legs... there was some of them that was connected with one another... i took all of them with a knife...
And there was the problem with the BIAS trimmer... i didn't know if the trimmer was not on the correct place i couldn't get any flanger... i was only getting clean tone, and when i started messing with this trimmer it came to life.
I am really happy with the effect, but i'm really anoyed with the clock noise... i haven't seen any original Electric Mistress so i don't know if there is any noise... i suppose it doesn't. I've got a friend that owns a Deluxe version... I don't know if it sound just like the 9V version, but i think it is a good point to start i'll ask him to lend me for a while.
Well, let me explain to you a little further so that you can help me!
I'm using everything with correct values as is in your project file. There is one resistor wich is 13K that i'm using 12k. This is the only substitution... but i don't think there is any problem with this.
I'm also using the output buffer volume drop compensation and it is working fine. I got it from your TZF project file.
I also did a led flashing with the rate of the LFO. I've made a transistor buffer for the led, so i think it is not the problem too... i also tested the unit with the led disabled but the noise is still there.
I'm not using switchmode power supply... i'm using a common power suply... i also tested with alcaline baterry, and with commom batery... but the noise was still there... i also tested with an fouth power supply, but with same results...
Does your unit have any kind of clock noise?
One thing that i noticed is that when the effect is in bypass mode, i can still hear some clock noise, but very very little... only when i put the amplifiver very loud... i tried to put shieled wires from input to 3PDT and from output to 3PDT, i think it helped, but just a little... the noise is very hearable (i don't know if this word exists)...
About your question: the frequency of the noise change at the rate of the LFO! But it also change when i change the rate and feedback pots... when i change the feedback pot is like a volume control... the noise gets louder the more i set the feedback higher... and when i raise the range pot, the number of clocks (pof, pof, pof) gets higher too...
Is there anything else i can do?
I don't want to give up!
Sorry for my poor english and thanks again for helping me!
Wagner.
Hey,
I fear that there are a couple of possibilities ;)
Since you mentioned that you had some shorts around the IC legs it might be that there is an additional short that causes bleeding of the LFO into the audio path.
Could you check with an audio probe (10µ or larger electrolytic cap with neg side to the tip of a guitar cable, gnd of the cable to gnd of the PCB) whether you hear the LFO at other spots: e.g. pin14 of the LM324, pin 1 of the 4558, junction of R1/C2 ?
Just take care when checking because if you touch e.g. LM324 pin 7 or 8 it will be damn loud.
Alternatively, you could try to add a huge (2200 µ or larger) cap between PS+ and PSGND.
Do you have the 200R resistor R5 or did you use a jumper?
BTW, nothing wrong with your english. I'm also not a native speaker and I can easily understand you :)
If someone else has some ideas what might be wrong please don't hesitate to chime in ;)
Peace,
Markus
Hello Markus,
I've made some tests and the results are:
I tried to check every conection on the board looking for shorts, but i couldn't find no more shorts.
I also tried a 3300u Cap between PS and PSGND, but no success. In fact i tried raising C1, C2 and C3... but the clock noise didn't went away.
I've made an audio proble too, but i didn't get any clock noise on the points you mentioned.
I'm using the 200R resistor, but i also tried without it with same results.
One thing that i noticed, maybe i'm saying something wrong, but i think the clock noise is coming from the GND.
I say it, because when i put the effect in bypass mode, i can still hear the clock noise (just a little noise in bypass mode, but i can hear it)... but in bypass mode the only connection with the rest of the board is the GND wire.
I took out the GND wire coming from the board and the noise dissapear in bypass mode.
I also notice that comparing my values with the table provided in the beginning of this thread, the value i get on the PS- point is -3,15V and the correct value (i assume) is -1,8V.
This is the electrical potential on the 200R resistor... i suppose my value is very high... i think my box is eating more current than needed (maybe another short?)...
Another point is that this value ranges from -3,15V to -3,02V depending on the clock (when the clock is high it starts at -3,15V and goes to -3,02V and when the clock is low it ranges from -3,02V to -3,15V)... i think it can explain the clock noise... maybe the GND is changing it's potential too...
I don't know what to do anymore... i think i'll try to build another board... but i've made two boards and they have the same symptons...
Is there anything else i can do?
Thanks,
Wagner.
Hey Wagner,
I' pretty stuck now. So I still hope that others with more experience than me will chime in :)
Anyway, I will continue guessing... ;)
To answer a question from your previous post: no I don't have any LFO noise. As far as I know others also didn't have any LFO noise...
Your last post got me thinking about gnd connections.
Since you already built two boards with the same symptoms it might be that the bug is not be directly related to the board.
Quotewhen i change the feedback pot is like a volume control... the noise gets louder the more i set the feedback higher..
This seems somewhow normal... besides the fact you shouldn't hear the LFO at all. ;)
What puzzles me is that you wrote "and when i raise the range pot, the number of clocks (pof, pof, pof) gets higher too..."
Does the LED also flash faster when you increase the range pot? Do experience this also with both boards?
QuoteI also notice that comparing my values with the table provided in the beginning of this thread, the value i get on the PS- point is -3,15V and the correct value (i assume) is -1,8V.
This is the electrical potential on the 200R resistor... i suppose my value is very high... i think my box is eating more current than needed (maybe another short?)...
Indeed, this seems a bit high. Doesn't leave much headroom...I would replace the 200R with a jumper.
Do you have this voltage on both boards?
Quotei think it can explain the clock noise... maybe the GND is changing it's potential too...
Good idea to check :)
A wild guess: maybe the ground connection between your Mistress and the amp has some resistance (faulty cable, oxidized jack in the amp, whatever). This might be a reason (again if anyone thinks that I'm talking rubbish please correct me! :) )
With the Mistress connected to your amp (power supply not connected to the Mistress, amp off) you might try to measure resistance between the "GND power supply" pad on the board and gnd at the input jack of your amp (or at the chassis of your amp).
Alternatively, you could try with power supply connected but amp off to measure DC voltage between gnd at output jack and again gnd at the input jack of your amp (or at the chassis of your amp). This way you could directly measure whether ground potential changes. Just might be that the changes are too fast to be seen on your multimeter.
Another question: did you connect the input jack's ground of your Mistress to the pad labelled "In Gnd" on the Mistress board? If yes you might try to connect the input jack's gnd, the "GND power supply" pad on the board and the power supply gnd to the output jack's gnd.
Finally another wild guess: if you happen to have another LM324 you could try to replace the one on the board. I'm not too optimistic but who knows.... ;)
Hope something helps......at least in finding the bug ;)
Peace,
Markus
Hello Markus,
sorry for a long time since i didn't put any information.
i couldn't solve the problem yet.
i'll put the effect on the breadboard to make further checks.
i think is the best to do right now.
i'm thinking about splitting the effect into 3 boards ( 1 for PS, 1 for LFO and 1 for audio path).
what do you think?
Hi Wagner,
I'm sorry that you were not yet able to solve the problem.
I can understand that you want to try a different approach with spliiting the board into three parts, especially since you really tried hard to find the bug.
Just would be a pity because you already did two boards and as far as I know there were already a couple of builds that used the layout without LFO noise. :(
Still hoping that someone with more experience than me will chime in and help ;)
Could you probably post images from the bottom and top of your PCB?
Might help....
Peace,
Markus
Hello Markus,
As you said, i also think it is strange that i have this problem as your board are already verifyed and a lot of people make it work wothout clock noise.
I'm sure it is a problem with my build, but i just can't find.
Well, soon i'll post some pictures and my wiring schematics too, maybe we could find something.
Maybe the effect is ok and there is other problems (for example: i live beside a radio station, so we can hear the station on the phone, and on my amps too... i'm not saying that this is the problem... i just want to say that could be some other external problems as the effects work very well except for the noise).
I still hope someone could help too... maybe someone who have a knowledge in clock noise.
Thanks for your help, and i'll post more info soon.
Wagner.
Hello Markus,
now i have a lot of info, so maybe it is easier to find any bugs.
i have some pictures, a wiring diagram and an audio file with the clock noise.
the first picture is how the effect is.
it has 2 3PDT... one for on/off and one for changing filter matrix.
(http://www.psychobox.com.br/em/em1.jpg)
the second picture is insite the box.
it is a mess... and i have the daughther board in my hands (it has the output buffer compensation and the led buffer, for led flashing).
(http://www.psychobox.com.br/em/em2.jpg)
the third picture is a look at the main board.
as you see, i'm using huge filter caps. hehehe.
(http://www.psychobox.com.br/em/em3.jpg)
a close look on the daughther board in the picture below:
(http://www.psychobox.com.br/em/em4.jpg)
the bottom side of the main board.
there are glue on the corners.
(http://www.psychobox.com.br/em/em5.jpg)
Inside the box... i tried to make this mess a little understandable.
(http://www.psychobox.com.br/em/em6.jpg)
My wiring diagram.
(http://www.psychobox.com.br/em/wiring.jpg)
the audio file with the noise clock.
the audio quality is very poor.
the mic is bad quality and it clipped a lot.
but you can hear clearly the noise.
http://www.psychobox.com.br/em/em.mp3 (http://www.psychobox.com.br/em/em.mp3)
Let's see if we find something strange now.
Thanks again for your patience with me!!!
That's very kind of you.
Wagner.
I have some ideas but I first have to ask. Did you first build, test & calibrate this before adding on the daughter board. And, please give a quick description of the calibration procedure you used.
Stay :icon_cool:...
Hello OldSchool,
Thanks for answering.
Yes, i first build and calibrate the effect and after i added the daughter board.
The daughter board has only an output buffer and a led buffer, so i don't think it can change the sound.
The clock noise is on the effect.
I calibrate the following way:
- bias trim-pot: set for lowest distortion
- feedback trim-pot: i set the feedback pot on max and then increase the feedback trim-pot until the unit started ascilating, then i decrease the feedback trim-pot a little bit
- clock trim-pot: i set for lowest clock noise. in my settings this trim-pot is on the minimum value, but i can still hear the clock noise, if i can solve this problem i want to increase this trim-pot because the effect is nicer for me with this trim-pot a little high.
If you need more information, please ask.
Thanks,
Wagner.
Hi Wagner, I noticed a couple of things in your pics & diagram and have to ask... Do you have the I/O shielded wires hooked up as in the wiring diagram (can't tell from the pic)? Common practice is to hook up only one side of the shield to ground (it's drawn w/both shields wired together at the switch). Next; I would shorten & "clean up" all the wiring a bit. The speed pot (LFO) wires would appear to be almost draped over the tip of your output jack. I'm not saying this is the cause, but there's no way anything good could come of that. These things should be tried first, IMHO. If the ticking (that's LFO) persists you can try this: Put a 47R resistor in series w/R29 (lift the V+ side of R29, connect one side of the 47R here, connect the other side of 47R to V+. At the junction of R29 & 47R connect the (+) side of an electrolytic cap (22uF - 47uF should do it); connect the (-) side to ground. This should help suppress the ticking. Also, you can try putting small (100n ceramic) PS bypass caps from (+) to GND on all the IC's. You can do this on the solder side of the board. While reading some of your posts I noticed some strange V's mentioned. Do you think you could post a list of actual V's from your unit like in the first post; using the same test conditions stated there? I know it's a bit of a pain, but this might help in some way.
Hang in there,
Dave
BTW; that's some beautiful artwork on that enclosure. :icon_cool:
Quote from: wcampagner on September 26, 2009, 03:55:41 PM
...
http://www.psychobox.com.br/em/em.mp3 (http://www.psychobox.com.br/em/em.mp3)
...
Well, it's not LFO ticking, because it has higher frequency. Plus this frequency is altered by LFO..
Could it be some clock drop-outs? Do you have oscilloscope by any chance? I'd check clock signals at pins 3 (or 14) and 8 (or 10).
Or something wrong with one or more cells in SAD1024 chip (ouch!) sending "black holes" thru the BBD line - but intervals (at the end of the clip) are to big for this probably..?
Or some mess from PSU - what PSU do you use?
Sorry, if my questions were answered before, I did not go thru whole thread yet.. :icon_redface:
T.
Quote from: wcampagner on September 26, 2009, 03:55:41 PM
Hello Markus,
now i have a lot of info, so maybe it is easier to find any bugs.
i have some pictures, a wiring diagram and an audio file with the clock noise.
Hey,
sorry for answering that late. I missed your answer.
However, glad to see that we are not alone anymore :)
Will work through it......
Peace,
Markus
One more question: where exactly did you connect the output jack gnd to the mainboard gnd?
I can't see from the pictures. Is the 200R R5 replaced by a jumper and instead you wired it like shown in your drawing?
Otherwise, I agree with what oldschool and Thomeeque suggested.
I also checked the mp3. Indeed, it's not LFO ticking but clock noise, or maybe a mix of clock and LFO?
A switchmode ps also came to my mind. However, Wagner already mentioned that he uses a conventional ps. :(
Really a cool artwork! 8) Reason enough to get it working....
Hello,
Thanks for all the answers.
I did a lot of tests during the last days.
But i still couldn't find the bug.
There must be a really stupid thing i'm doing wrong but i couldn't find yet.
Let's go for the answers:
I have the IO shieled wires the way it is on the wiring diagram.
I had it only on the input, but i also put on the output... just to see if the problem was there.
But it wasn't, so it doesn't matter.
I've made a cleanup on the wiring too as suggested, but no difference.
I've tried the trick with the 47 resistor and capacitor, but no difference too.
I've tried it in most of the IC on the board.
I've put 100n ceramic on all IC's too, no way too.
Maybe there is something wrong with the SAD1024 as Thomeeque said.
I have two of them here, i bought from a french store... but both have the same sympton.
Is there a possibility that they are damaged?
I changed all the IC's except the SAD for new ones too... no way too.
I don't have anymore SAD's here, but i could find someone selling a MN3007... maybe i'll buy it and try with this one as suggested on the OldSchool retrofit thread.
OldSchool, does it sound like with SAD1024?? I think i'll try it to compare... what do you think??
Tomeeque, i tried with 2 PS and 2 battery too... all with same results.
Markus, about the output jack, i also tried several places for the GND... but now it is like the wiring picture above.
The 200R: i also tried with and without it... didn't change the noise.
I think the problem is a mix between LFO and clock noise too.
My board with values is below, is there something wrong?
(http://www.psychobox.com.br/em/em7.jpg)
As you see, the values are basicaly a bit lower than the table in the first post.
It sems to me that my board is consuming more current than the one in the first post.
I also noticed that the V+ is changing from 6,46 to 6,74.
I think it is strange because it shouldn't change.
As is in the PS- (where the 200R resistor is).
They are both changing.
Anyway, after some time, i put the effect on the breadboard.
It was easier to make all the tests.
I was thinking that it should be the LFO making the V+ oscilating.
So i put the audio path in the top part of the breadboard.
And the LFO on the bottom part.
The problem was still there, so i decided to use two PS's.
One for the audio and one for the LFO.
The was a hum on the amplifier with this configuration, but the PS in the audio path was not oscilating anymore.
But the noise was also there.
That's why i'm thinking there is a possibility that the SAD is damaged.
In the beginning i was thinking that the problem was because the PS was oscilating.
But even with a non oscilating supply the noise was still there.
Well, i'm almost giving up... :icon_cry: :icon_cry: :icon_cry: :icon_cry:
Very "SAD"...
I'll give my last try in the mn3007.
And any sugestion you very kind people gives me.
But i just don't know anymore.
Thanks for the compliment about the enclosure.
That's the way i do all my pedals.
There is more models in my web site: www.psychobox.com.br
There is only a few pictures, i'll put more and sounds too as i have more time.
Thanks a lot for all your help!
Wagner.
i forgot to mention:
All the tests i took out the daughter board, so there is no more outpt buffer and led buffer, ok?
Thanks again.
Hey,
sorry to read that you were not able to solve the problem so far. :(
I don't have any experience with damaged SAD1024s. Since you tested two and both give the same symptoms I'm tempted to believe it's not the SADs.
At least for me it's hard to believe that both have the same damage. But as I said. I don't know how damaged SADs sound.....
Regarding voltages: as you mentioned it seems your board draws more current. According to the SAD data sheet it can be run with as low as 5V supply.
On your board the SADs see 6,74-2,06 = 4,68 supply or less. Maybe this is already too low?
Have you tried to jumper the 200R and the 10R R1 resistor? This should give you 9V supply on the SAD.
Regards,
Markus
Hang in there everyone who is debugging! Took me two days to track down all my errors, but it works! Turned out the jumper grounding the sad1024 was not making contact to the ground. I also made a few other bonehead mistakes. That's what happens when you populate the board at 2am while listening to the new AIC record. wcampagner, is there a schem for this daughter board of yours? Sounds like a pretty good idea! I haven't totally settled on TBP or bypass yet. I'll have to see how much it effects my setup.
hello,
thanks for the answers.
Markus,
yes, i tried to jumper both the 200R and the 10R resistors.
but no way to cure the noise.
Jmascis,
i also checked the ground connection with the SAD, but ii was ok as i'm getting flanger.
the problem is only the noise.
anyway, i got my hands on a MN3007 and did the modifications as discussed on the retrofit thread.
i got it working... the flanger sound is okay... as OldSchool said, it is just like the original or better.
i like the sound, but the noise was there too... :icon_evil:
well, there is a good point that i think my two SADs are not damaged.
and the bad point that it is not working.
i'll take a break now.
Put this effect aside and do other things.
i'm tired and after some time i'll be back to try to find the bug.
i'm thinking of building an uni-vibe... maybe this one is easier...
thanks a lot for all the help you gave me!!
the daughter board schematic is below.
if you have any doubt please ask.
(http://www.psychobox.com.br/em/mistress.jpg)
Thanks a lot!
If i have any success i'll post here!!
Quote from: wcampagner on October 06, 2009, 11:16:54 PM
...
and the bad point that it is not working.
...
Maybe you should try to take your build somewhere far from your place, maybe there really is some strong electromagnetic disturbance in your area??
Quote from: wcampagner on October 06, 2009, 11:16:54 PM
the daughter board schematic is below.
if you have any doubt please ask.
IMO V
ref network is missing at your schematic:
(http://thmq.mysteria.cz/diystompboxes.com/wcampagner_emdb_vref_fix.gif)
T.
Hello Thomeeque,
I'll try to take it far away from home and test it again.
I live beside a radio station, so i always have problems with really strong interference at home.
Maybe this is the problem... i'll let you know.
Thanks for the corrections on the daughter board too.
It is working allright without the Vref network.
Is there any problem leave it this way?
I also think the 100k trim-pot on the output buffer is very big.
With just a little change the volume increases a lot.
I think i'll switch to 10k maybe.
The 100uF capacitor on the LED blinking buffer is too strong too.
I ended using 10uF.
This way the LED does not blink in the fastest settings.
It seem like a heart beat, nice effect.
I'd like this effect on all settings, but the slowest settings are just too slow for the capacitor discharge.
Thanks,
Wagner.