adafruit industries

asmodee · Post by **asmodee** » Sat Mar 04, 2006 12:32 am

VCA testing as described in the fabrication manual results in the desired output on VR7 (rightmost pin) when the gate is connected to R146, however it does not decay, which is what I think it should do.

I tested the BA6110 as described here: http://wiki.jonnay.net/x0x/vcatesting

My pin 4 shows 1.0v when gate is not present and 1.2v when gate is present, although my scope isnt calibrated so im just guessing. It goes up very slightly when the gate is connected.

Pin 9 shows 12v at all times, gate or no gate.

Anyone else run into this? Could it be another bad batch of BA6110s? I don't know what the actual problems were with the other date code.

I probably have a solder bridge somewhere that I haven't tracked down yet ;P

asmodee · Post by **asmodee** » Sat Mar 04, 2006 3:14 pm

Ok I found a problem, I forgot a capacitor (I think it was C41). Now it appears as though the envelope effects the VCA, but it doesn't appear to go completely silent.

I misunderstood what pin 9 is supposed to show. It should always be 12v regardless of gate.

Either way that pin 4 oddity is still present even after I put in the missing cap. With no gate the voltage at pin 4 is 1.017 and with gate is 1.133. I suspect this is why the VCA doesn't go completely silent.

I went through and ensured that every part I've put in so far is in there correctly. I went over all the solder points again and have checked for solder bridges.

guest · Post by **guest** » Sun Mar 05, 2006 4:02 pm

does your gate go off completely
check the d34 and d35 juncture
it should be at 0v when no note is playing

asmodee · Post by **asmodee** » Sun Mar 05, 2006 4:11 pm

It goes to 0.01v at that junction with no gate present. if I connect GND to gate input (R146) then the junction of R146 to R144 goes to 0.003v.

guest · Post by **guest** » Mon Mar 06, 2006 3:42 am

that is good
can you check the voltage at
the juncture of d27 an q3
it should be 0v with no gate present

if it is 0v
then there is most likely a problem with
q1 q2 q3 or q4
you can test this by removing q1
if the vca still doesnt go quiet
than there is a problem with the vca
or the trace to pin4 has some slight
conductance to a higher potential
you can hit the solder joint with some solder wick
and deflux the area and see if that helps

asmodee · Post by **asmodee** » Mon Mar 06, 2006 1:02 pm

At junction D27 - Q3 the voltage is 0.45v with no gate and 0.56v with gate.

Additional tests:
Q32 - R132 junction with no gate is 1.7v with gate is 0.14v
R132 - Q31 junction with no gate is 1.8v with gate is 0.14v
Q31 - R131 junction with no gate is 2.2v with gate is 0.69v

Does this mean my problem is somewhere between D27 and D35/D36?

guest · Post by **guest** » Mon Mar 06, 2006 1:31 pm

it sounds like q31 is in backwards
or it is broken
try pulling it and see if that silences the vca

alternatively you can pull r123
but if q31 is bad youll need to pull it anyways

asmodee · Post by **asmodee** » Mon Mar 06, 2006 1:54 pm

Q31 wasn't in backwards, but I removed it. Now pin 4 of the BA6110 with no gate is 0.998v and with gate is 1.004v.

Now when I test the output of the VCA (at the rightmost pin of the volume) I see about 18mV of ripple when i have all of the knobs (except tune and volume since its not installed) turned fully clockwise.

So the VCA is now nearly silent with no gate, but it's control input (pin 4) is still not at the right voltage. Also when I connect the gate, the VCA voltage peaks and then decays as expected.

Should I try a different 733 in Q31 or is there something else that's the problem?

ironpyro · Post by **ironpyro** » Mon Mar 06, 2006 2:17 pm

Is there a way to test the BA6110 without having it in the x0x circuit?

guest · Post by **guest** » Mon Mar 06, 2006 4:55 pm

this is quite perplexing
if q31 is still out
can you try removing d27
basically that will eliminate any current sources
into the current mirror
so if it still goes on
then its either the current mirror
or the ba6110

asmodee · Post by **asmodee** » Wed Mar 08, 2006 2:20 am

ok I removed D27 and the voltage at pin 4 on the BA6110 is still around 1v regardless of gate. Although the voltage on pin 4 was not right before I removed D27 (but had Q31 removed) the output (measured at the rightmost pin of VR8) seemed to be nearly correct. It jumped up when the gate was applied and decayed until the ripple was around 18mV. Another oddity that I notice is that the DC offset of the right pin on VR8 drifts when i connect my meter or scope. It drifts faster with my meter. I suspect this has something to do with the load of my meter/scope affecting the opamp.

So since it seemed to be close to working without Q31, but with D27.. what does Q31 do?

I don't think Q31 was bad because after I removed it I tested it and it had a beta of 269. However the ECB seems backwards from the data sheet...

guest · Post by **guest** » Wed Mar 08, 2006 1:12 pm

q31 controlls the current into the vca
if its pinout is somehow backwards
that would definitely make it not work
although id be suprised if only one of the
transistors you recieved was faulty
in that particular way

although 28mv of signal doesnt seem like much
you will still probably be able to hear it in your mixes
id suggest pulling q1 and seeing what the pin4 voltage does

you can also check the trace from pin4 to q1
it might have a small short someplace
or perhaps there is an excess of flux or solder
the smallest current into that pin can turn the vca on

asmodee · Post by **asmodee** » Wed Mar 08, 2006 4:08 pm

you are right in that Q31 wasn't the only transistor with a pinout that is backwards from the datasheet. All of the 733s are backwards from the datasheet, but I assumed that the image on the x0xb0x pcb is showing the correct orientation as that is how I put them in and everything tests correctly up to the VCA.

If I'm reading the schematic right, the signal going into pin 4 on the VCA can be traced back to D27 and Q31 and prior to that the signal to Q31 can be traced as far back as D35 and D36. The signals at D35 and D36 are correct. So if I replace Q1 and Q31 and leave out D27, yet I still have odd values at pin 4 on the VCA this would lead me to believe that the issue is between D35/D36 and Q1. Is my thinking correct?

adafruit · Post by **adafruit** » Wed Mar 08, 2006 4:12 pm

yeah theres this wierd thing where some 733s are not the same pinout. this was a problem when we first started because i trusted the datasheet

but the x0xb0x schematic and board work with the transistors you've got..the 2sa733s are from mouser so maybe they have a datasheet up. i think fairchild makes them?

asmodee · Post by **asmodee** » Wed Mar 08, 2006 4:19 pm

I think I'm going to remove all the transistors and caps (between D35 and D36) from the VCA, measure all of the resistors in the VCA that are left (to make sure they are good) and then replace the transistors and caps with all new parts.

Or is that a bad idea?

adafruit industries

BA6110 oddities

BA6110 oddities