Mad Professor Snow White Fail

[nickbungus]

Hi Duckster, long time no... whatever the word is in forumish?

I've given up on this for a bit as I've got some jobs on that pay me rather that the other way round but I am going to give this one more go soon.  I think my problems were:
a.   I'd probably made a mistake somewhere that I couldn't find
b.   I was using some ebay cheap copper clad board and kept getting lifted pads and traces etc.
c.   I'd done some sloppy soldering on the pots ( and probably everywhere else)
d    I'd probably made a mistake somewhere that I couldn't find

I can repost my voltages if you think that would help but my last lot seemed to match Yazoos working voltages pretty close.

My plan is to go again with 100% new components and using some better quality copper clad I've just bought.

[pillo10]

My voltajes are posted too. In the pin 7 I have 1,81. As I said I bought new main componets (LM324, LM13700, 2N5457, BC550C), but the problem did not disappear.

[duck_arse]

thanks, jim, but I did say "limited". baffling.

and the diode in series with the transistors means pillo's base voltage of [erm, lost it, over the page] something less than his pin 7 is correct?

and pillo - the Q2 emitter connects hard to the Q1 base, so those two voltages should measure the same ........ strike that, I see it now. maybe check the value of that R13.

[anotherjim]

Well, you have an inverting amp to pin 8 with unity gain from the +V connection. The differences in pin 8 voltages are I think because...
A: Vref is what it is from the divider elsewhere.
B: The op-amp drives the output down as far as it needs so that pin 7 at the junction of R19/R9 = Vref. The actual values of those resistors must differ enough so the output doesn't have to go quite as low to equal Vref OR Vref is a little higher on that example OR a bit of both.

You might expect the output from the rectifier to change it a bit, and R23 is small compared to R9 & should give a lot of gain going into pin 7. But it's a mixer, and a virtual ground mixer at that so the paths via R19 and R23 do not interact. As it happens, with no signal, the rectifier output is close to Vref, so there is no difference between the input pins 5 & 6 from the R23 path. If you removed the +V feed to R19, pin 7 should sit at Vref.

I'm not seeing evidence of diode drop around the rectifier in the readings. I think it's all hanging at Vref due to leakage & the diodes are not properly conducting, you have to wait until enough envelope signal is present to see diode action.

[pillo10]

Thanks duck_arse

I checked the R13 and it is correct 330.

Sorry anotherjim. My electronic knowledges are limited and I don't know what you say
and I have to traslate it to Spanish and it is worse, but thanks

Could it be a desing flaw in the PCB?

[anotherjim]

Pillo10, your 1.8 volts on LM324 pin 7 may be because you have a little voltage difference on pin 8 compared to pin 5 - Check voltage at a point between R23 and the diode D1. Make sure sensitivity control is minimum resistance when taking measurements and the input is silent. Pin 7 volts should ideally be less than 1.2v when the input is silent, so the filter sweep will be greater when you do play.

[pillo10]

Thanks anotherjim

I have been taking information and in some webs is recommended to change some components (D5 1N4148, R18 47K, D6 1N2A4 and C8 1uf) because them improve the sound. I do it and it works same, only hear the buzz.

I have done that you have said anotherjim. I take the volt. between the R23 and D1 and it's 3,99 v.
I take a new lecture in LM324 pin 7 and when the Sensitivity is minimum it is 1,83 v. and when it is maximum 0.95 v.

I take new lectures with the changes without operating swich and operating it:

LM3234
1: 3,95
2: 3,95
3: 3,95
4: 7,95
5: 3,96
6: 3,96
7: When the Sensitivity is minimum 1,83 and when is maximum 0.95
8: 3,79
9: 3,95
10: 3,96
11: 0
12: 3,96
13:3,96
14:3,96

LM13700
1: 1.2
2: Here the lectures are rare. First I take 1,2.............actioning swich it goes to 0,02 v. but little by little...and then, without actioning it stays in 0,02
3: 4,01
4: 3.96
5: 4.57
6: 0
7: 4.45
8: 3.80
9: 4.30
10: 5.47
11: 7.95
12: 5.47
13: 3.95
14: 3.95
15: The same than pin 7
16: 1.20

2N5457
D: 7.95
S: 4.05
G: 3.59

BC550
C: 4.95
B: When the Sensitivity is minimum 1,70 and maximum 1.20
E: 1.2

BC550
C: 4.95
B: When the Sensitivity is minimum 1,66 and maximum 0,80
E: 1

[pillo10]

May the mistake be in the electrolitic capacitor 100 uf conections of my PCB?
Please may you verify it with the PCB and electronic diagram posted in the first page? I think that it is wrong

Thanks

[antonis]

Sometimes we tend to overcome "basic" malfunctions..
(like loose connections between transistor legs and sockets..)

[pillo10]

Confirmed. It was the incorrect positioning of the capacitor C7 in PCB. The + of capacitor has to go in V + (red line of PCB).

It sounds great.

thank you for all your help

[anotherjim]

Congratulations, and...
"I have done that you have said anotherjim. I take the volt. between the R23 and D1 and it's 3,99 v.
I take a new lecture in LM324 pin 7 and when the Sensitivity is minimum it is 1,83 v. and when it is maximum 0.95 v."
That confirms pin 7 is ok - except - I would expect the other way -  1.83V with sensitivity maximum!

[pillo10]

Thanks anotherjim

[Kipper4]

Pillo can you post your final voltages ? It might be helpful to Nick.

[pillo10]

I think that Nick had my same problem because It's the same PCB. But if he needs it, I will take new reads 

[nickbungus]

Hi

Are we saying the actual layout was wrong rather than a mistake on your behalf Pillo?

I hope not.  I threw my Tagboard Effects board in a fit of rage, punched it, kicked it, did a poo on it and then burnt it.  Really I just binned it but I dont have it.

I still have my RevolutionDeux board.  Which one did you use Pillo?

[pillo10]

Yes It's. The layout is wrong. The 100uf electrolitic capacitor is bad conected. It has to go, the + in the red line of the layout. Only modifing that your snow white will sound grate

[anotherjim]

Good catch!
On the pcb layout Nick posted at the beginning of the thread, that 100uF appears to have it's + wire on a track going to the LM13700 instead of the +9v power.
Well found!

[still4given]

deleted

[Blitz Krieg]

this is related to nordinator's fix?

http://www.diystompboxes.com/smfforum/index.php?topic=113861.msg1060117#msg1060117

[chemosis]

i know this post is old but ive built  now white that sounds the exact same as in the video. it doesnt hae a strong enough wah sound. its barely there so i built another that wrked great a few months back. then i sold it. now ive built another one this week and it dosent hae a solid wah sound like the first. almost as if it might have to little sense or to much sensitity, but it seems like not enough maybe. it sounds liked a parked wah. itss mosly clean signal