BL3208

[dizso]

Hi!
I am working on Aquaboy with chip BL3208. And it is very distorted. It is ok until the input of the BBD, but the output of the BBD is very noisy and distorted.
I tried the next:
- biasing
- reduce the input level before the compander
- reduce the input level before the first filter transistor
- adjust the symmetry after the BBD
If the clock generator is reduced to minimum and there is no delay, it is not noisy and distorted...
Any idea? Thanks for any help!

[dizso]

Some more experiment: just at the output of the BBD it is very noisy even in the best position of the "cancel" pot. In the deep frq it is not very ditorted, but the heights (the middle of the E1 string) is just a hash,  not a real sound.

[madbean]

What are your voltages on the BBD and clock pins?

[dizso]

It is 8,98V at vDD.
Which else can be interesting?

[dizso]

so, now i could have measured some voltages:
BL3208
pin 8 (Vgg) - 8.2V
pin 5 (Vdd) - 8.8V
pin 7 (In)   - 5.7V
BL3102
pin 1 (Vdd) - 8.8V
pin 8 (Vgg) - 8.2V

[madbean]

Those voltages look pretty good actually so I don't think that is the problem. Just to be clear: does the noise and distortion carry through to the  Mix output of the delay? IOW, is it isolated to just the BBD output or are you hearing it mixed in with the dry signal at the output? It's not necessarily unusual to have some of what you are describing at the BBD outputs but it should not carry through past the de-emphasis filtering and expandor. Also, if you have another BL3102 it wouldn't hurt to pop it in and see if that takes care of it. Or, you could use an MN3101 if you install D3 (just going through all the debugging things I would try here).

[dizso]

The distortion is somewhere in the circuit, the dry signal is muted.
I tried with two BL3102, but i have not tried with MN3101. I need to find one...
Do you think, the clock can cause the problem?
Before the BBD the sign looks good (by ear), after the BBD is distorted, the higher voices more.
The clock was measured by scoop: it varies between 10-100 kHz.

[madbean]

That clock range seems very wrong. If 10kHz is your lowest setting then that could be the problem. The proper clock setting on a DM-2 when biased correctly usually happens between 6 and 7kHz.

Can you take another look at C33? Are you positive you used 100pF there?

[dizso]

C33 is 100p silver mica. So hopefully it works properly.
According to the data sheet the BL3208 waiting for 10-100 kHz. So why it is wrong?
Should i reduce C33?

[madbean]

You know what I think you are right. The clock frequency range is different b/w the 3101 and 3102! But, no I don't think C33 should be changed. Let me look through my builds and find the Bl3208 version I built. I'll see about getting you some better reference to go from.

[dizso]

Thanks for your help!
No hurray, i am busy with my work during the weekdays, i can work on the delay at the weekend...

[dizso]

Hi Measbean!
You may find something about the BL3208 issue?
Best, dizso

[dizso]

So finally I have changed the BL3102 clock to MN3101. The BL made a crazy clock sign...
This way it works nicely! Problem solved.
ps.: maybe the BL3102 was fake or I do not know... I have tried two of them from the same stock.