aquaboy 2013 bebug help-no delay

[krankor]

hi,

using a mn3205 (checked the jumpers already!) here are my voltages:

IC1 LF353
1:4.5
2:4.5
3:4.5
4:0
5:4.5
6:4.5
7:4.5
8:9

IC2 NE570

1:1
2:1.8
3:1.8
4:0
5:1.8
6:1.8
7:3
8:1.8
9:1.8
10:3
11:3
12:1.8
13:9
14:1.8
15:1.8
16:0.8

IC3 MN3205..COULD IT BE FAKE?

1:0
2:1.5
3:0.28
4:0.3
5:5.6
6:0.9
7:2.7
8:4.9

IC4 3101
1:5.6
2:0.9
3:0
4:1.5
5:1.5
6:1.5
7:3.8
8:4.9

Q1
E:3.4
B:3.8
C:9

Q2
E:2.1
B:2.9
C:9

Q3
E:.005
B:0.288
C:9

Q4

E:0.005
B:0.005
C:9

probe shows that the signal gets very low at the junction of C35 and R35. should the voltage coming out of pins 3 and 4 on the 3205 really be on the order of tenths of a volt? I would say that maybe the voltages for Q3 and Q4 looked funny if it werent for the fact that the output of the 3205 seemed so low.

any help appreciated!

[krankor]

sorry meant aquaboy *2012*

[Scruffie]

Your 3205 could be fake, as could your 3101, where are they from? Got a photo?

Both of which are only getting 5V supply which isn't right and for those low voltages the clock still seems off...

Remove your 3205 and remeasure its pins and the same with the 3101 just to rule out either being fake and pulling the voltage down as your OpAmp voltages are fine, not checked your compander yet but at a glance it's okay.

After that, just start following 3205 pins 5 and 3101 pin 1 and see where you're loosing voltage.

[alanp]

v3205d's do not sound flash compared to mn3205's, but it could be worth getting a pair just as known-good test devices.

[krankor]

cool thanks for the replies. The chips are from chinese sellers on ebay:
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=150720096426&ssPageName=ADME:L:OU:US:3160

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=180476363561&ssPageName=ADME:L:OU:US:3160

i bought both of these chips about a year ago and the dealers still have stores on ebay so at least that is mildly encouraging. sort of.

oh wait i also bought some 3007s and 3101s from somewhere else at one point and i know the 3007s work. well it has been a while..not sure. will try to get some pics up tomorrow.

also should have mentioned that i etch my own boards, so, you know..

[Scruffie]

It's more of a power issue, the chip seems to have near enough the right voltages apart from them being so low (and the clock being wrong but we'll put that down to power issues).

Do you have a link to the project?

[krankor]

A lot has happened...but still no delay! The 3101 turned out to be fake or i fried it, or both but i forgot i had one lying around from a different seller and hey it worked! nice healthy clock pulses on the scope and here are the new voltages:
IC1 LF353
1:4.5
2:4.5
3:4.5
4:0
5:4.5
6:4.5
7:4.5
8:9

IC2 NE570

1:1
2:1.8
3:1.8
4:0
5:1.8
6:1.8
7:3
8:1.8
9:1.8
10:3
11:3
12:1.8
13:9
14:1.8
15:1.8
16:0.8

IC3 MN3205...now thinking it is fake or fried
1:0
2:4
3:3.7
4:3.7
5:7.9
6:4
7:3.8
8:7.2

IC4 3101
1:7.9
2:4
3:0
4:4
5:4
6:3.9
7:3.9
8:7.2

Q1
E:3.4
B:3.8
C:9

Q2
E:2.1
B:2.9
C:9

Q3
E:6.3
B:6.9
C:9

Q4

E:2.1
B:2.1
C:9

When i look at pins 3 and 4 on the mn3205 on the scope all i see are clock pulses. is the delayed input waveform supposed to be riding on those? they then go through the cancel trimmer and it does its job. then all you see are tiny spikes which i am guessing are artifacts from clock pulses that didnt quite cancel. but no input waveform.

so with the functioning 3101 here are voltages for the 3205 pins with the 3205 removed:

1:0
2:4
3:0
4:0
5:8
6:4
7:3.5
8:7.2

and the 3101 removed (3205 back in)

1:8
2:0
3:0
4:0
5:0
6:0
7:0
8:7.7

couldn`t find the aquaboy 2012 project on this site so i included the pdf and a pic of the mn3205. ok i think that is everything. thanks again!

[Scruffie]

Can't really tell much from the photo of the 3205, doesn't look clearly fake but that don't mean much.

Your voltages are looking a lot better now at least.

If you audio probe, does signal actually reach the 3205 input?

[Govmnt_Lacky]

Did you take the picture of that MN chip in a mirror? The text is backwards!

[Scruffie]

Did you take the picture of that MN chip in a mirror? The text is backwards!

Haha, reminds me, I just repaired a guys chinese DM-2 clone which uses real MN3205 and my first debugging stop after voltages were 'well that BBD is backwards!' so I stuck in a bl3208 the right way and that worked but after speaking to the guy, he said it had been working before and he'd never touched the chips... so I put it back in how I found it... and they'd re-stamped it upside down and put the notch on the wrong end!
Makes me wonder if in future getting ebay chips I should try them both ways round... just to be sure.

[Govmnt_Lacky]

Makes me wonder if in future getting ebay chips I should try them both ways round... just to be sure.

That is just wrong on so many levels!!!

Not what you are suggesting but, the fact that you would actually have to try it! 

That is one good way to ruin a good chip and never even know that it was a good chip until it is too late!

Oh well... the risk you assume with the eBay deals!

[Scruffie]

Makes me wonder if in future getting ebay chips I should try them both ways round... just to be sure.

That is just wrong on so many levels!!!

Not what you are suggesting but, the fact that you would actually have to try it! 

That is one good way to ruin a good chip and never even know that it was a good chip until it is too late!

Oh well... the risk you assume with the eBay deals!

Hence why i'd only ever test in a known working circuit against a working chip and do so as a 'last resort' but a BBD isn't quite as easy to kill as an OpAmp... probably going to be noisy after being backwards but if you're quick, ya might get lucky and get some use out of it. Anyway, most likely a one off but still I was a bit confused by that one!
*warning!* I am in no way suggesting to those with no experience of BBDs to just try swapping them backwards, you may very well destroy your perfectly good chip, I would only undertake this when all other tests had been exhausted as it appears to be a one off occurrence *warning!*

[krankor]

hehe! yeah sorry i took that terrible pic with my computer camera and it took way too long to even get that good of a shot of it. i wish someone would explain why you get a mirror image and why it would be so hard to correct for. weird. but anyway yeah to look at it you would say it looked pretty legit. and it might be-i made this thing almost a year ago during a real solder binge and i have been known to not think and just blindly plug in the chips without thinking about which way they are supposed to go cuz i am in such a hurry to hear if the thing works.

yes there is a signal at the input of the 3205. unfortunately haha! that`s what makes me think it is either bogus or i ruined it. spensive mistake if that`s what i did.  Thanks for the help!

i did run across this from Brian in another thread :
Okay, try this instead: remove D2 from the PCB and put a jumper in its place. Now put D2 (or another 1n914 or 4148) between pin1 and pin8 of IC4 on the bottom of the PCB as illustrated below in orange. This will set the Clock at the same voltage as the BBD and change the Vgg voltage to about 14/15 of the Vd supply.

If that doesn't work, it may be a bum BBD.

[Scruffie]

Your Vdd & Vgg are fine though... not sure what Brian was referring to in that quote.

So you have a clock signal, you get audio in to the BBD, what about after the BBD, how does that probe?

Don't forget to adjust your bias too!

[krankor]

Sorry for the delay replying. see what i did there? delay? funny shtuff :^| ...anyway. Probing at the output of the BBD you can hear a mixture of clock signal and input signal. The clock signal is filtered away by the time you get to Q4 and is healthy sounding until it passes through R33 where it sounds heavily attenuated and low pass filtered like it is coming from the apartment next door.

looking at the signal on the scope you see  the clock signals and what i am guessing are discrete voltage samples- square wave and higher amplitude spikes, all positive going. then  at the output of the compandor the square wave clock signals are gone and the spikes are both positive and negative going. after R33 it is hard to distinguish signal from noise.

i checked all the R and C values around IC1B and they are correct.