x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Mon Oct 01, 2012 8:03 am

Hello everyone,

I got a x0xb0x kit from a third party in germany, just finishing it up but of course going through some trouble first :)

I am an electric engineer, so I do understand the schematic and the circuits used pretty good, although, since I didnt grow up with the analog transistor stuff I really having a hard time understanding some of those circuits, there I could need some help.

My x0xb0x wasnt producing any sounds I thought a couple of days ago, no as I kept pressing my fingers on possible cold solderspots etc ( basic first try debug, I would call it :) ) I realized that I will get the wishes sound out of the headphones whenevr I pressed my thumb on the headphone out opamp and the VR4. I also have variable voltage sourced and an oscilloscope available for some more serious debugging later on :) but i am coming to that in a second.

Now that I knew my sound is working but is getting stopped somewhere later I did probe a lot on the PCB with the results that my VCA BA6110 pin 4 ( the control input current ) is not getting any signals ( no voltage and therefore no current i thought). Firts I did wanna buy a new BA6110, but I am not really sure if it really is broken or some other circuit in front of it. So I did simply attach some external voltage source to see if the VCA will work when i force it to do so. And indeed headphone output is noise free, everythign seems to be working like it is supposed to do. But my voltage attached draws about 10 mA, and in the BA6110 it says the control input max current should be 0,5 mA, so I guess my workaround works for some minutes but not forever.
And now you guys come into play, where could the error be and how can I realize a current source in bipolar?

I think my problem is the current source built by Q1,Q2,Q3,Q4, but I do not really now how it should work, perhaps you will know, or even know my error.

I think my BA6110 is correct but the transistors around it do not work properly since I can hear clear sound once I power the controlinput ( Pin4 ) manually.

Thanks for your help.
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by antto on Mon Oct 01, 2012 8:22 am

you've placed the ba6110 in IC15B right? not IC15A
We are here too: irc.freenode.net >>> #x0xb0x
..:: c0nb0x v1.00 ::.. http://forums.adafruit.com/viewtopic.php?f=7&t=29282 (new control app - win/osx)
VCO Tuning: http://forums.adafruit.com/viewtopic.php?f=12&t=24759 (do it the EASY way)
User avatar
antto
 
Posts: 1239
Joined: Thu Apr 15, 2010 2:21 pm
Location: 127.0.0.1

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Mon Oct 01, 2012 10:22 am

Yes, as I described, everything works fine, once I put some external voltage source on pin 4.
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by MMM on Mon Oct 01, 2012 3:36 pm

Please measure the voltage on the following diodes when playing a note in manual mode:

D34, D35, D36, D37
MMM
 
Posts: 420
Joined: Tue Apr 24, 2012 12:51 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Tue Oct 02, 2012 6:55 am

Ok, I checked them voltages, and its 0 V on every cathode.

I was suspicious about the GATe signal coming from the µC, is it supposed to be high or low? It is low in my case.

I did try to manually pull Q37 base to high voltage, then I had a kind of short circuit on that pin ( voltage supply got into current regulation while 0,8 V output and 15 mA ) so I now decided that my Q37 is broken , I did disconnect the Base input from the µC before testing, so I can be sure there is no current flowing into the µC.

What should the voltage levels be?

As I posted earlier i am not that familiar with the good old transistor circuits, still trying to understand those npn pnp and vice versa circuits built by Q1-Q4, I guess it should be a current source right?

As I am just looking on the schematic, I would say need 12 V or a high signal on D35 in order to activate the Q1-Q4 circuit, is tat right?

So right now my next guess would be to somehow activate( close) the Q37 pnp transistor.
But right now it seems to be broken, I will short E-C next step and see if the Q1-Q4 will be activated, so a control current is flowing into my BA6110.

Thanks for showing interest :)
Greets
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Tue Oct 02, 2012 8:38 am

Update:
I did power up the Gate with 3 Volt again and do see a high signal ( 12 V ) at D35, D36 and D34. NExt I do understand that Q32 is activated , then I dont get the circuit any more.

There is a 1M5 in series with 200 Ohm, so once Q32 is activated, there is no chance that Q31 will also be activated, is that true?

right now I would state, in order to function right, volatage on D36 must be low, and voltage on D35 must be high.

basically this is something I tried to simulate by putting a high signal ( voltage source ) to the anode of D27 ( this is also the base signal for Q3 and Q4 ) , by swithcing those 2 npn transistors on a current can flow into Pin4 of the BA6110 and music is produced.
Problem still is, there is also no clear high signal needed, when I attach 0,6 - 0,69 V it wokrs, when getting lower it shuts off, when getting higher the current regulation pops in, and I think it would take way mor ethan 30 mA to be in voltage source mode again and I do think this would already destroy some circuit on my pcb.

So another principal question: In my opinion, D36 and D35 are always high and low together, because of the similar circuit in front of them, both connected to the same GATE signal, coming from the µC, or do I understand it wrong( could definetly be )7


Update#2:

Sorry , I do meassure 12 V on the Diodes when playing a note, at least I think that is, I did only measure the Pin4 Input when playing actually a note ( I did only now realize that Gate Pin of the µC might change when pressing a keyboard button :) dumb me )

so I do get 1 V at Pin 4, but still no sound, when I do attach a voltage source on that Pin it kind of works, as I told before, so why the hell is it not working with the in built circuitry but with an external source, I do not really get it??
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by MMM on Tue Oct 02, 2012 10:42 am

A little bit of food for thought:

The envelope is built out of two parts: a stationary part and a changing part.

Through D35 passes a Gate signal, but through D36 just a trigger signal. I'll look into this a bit more deeply later. About 1V on pin 4 of the BA6110 seems about correct, though.
MMM
 
Posts: 420
Joined: Tue Apr 24, 2012 12:51 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Tue Oct 02, 2012 10:55 am

Hello again, thanks for this insight.

I am just meassuring all the transistors with my scope, as I dont understand the pnp nor npn ciruictry really well, this already helped me a lot.

I do emasure while pushing the keyboard, producing sound though.

I also meassured out thos trigger and cnostant signals and I have no clue right now, perhaps I do not have a problem with my VCA at all.

But perhaps this is strange too, since I get a perfect signal once I put an external source on Pin4 of BA6110. Perhaps someone could tell me what do I have to expect at the signal out of the BA6110 ( Pin 8 ) I do think I should meassure my sound signal there, no?

Because I get a 700 mV dc signal there and nothign else, so it is always blocked from the C behind it :(

I wil let you know once I cracked it :)

Thanks anyway
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by MMM on Tue Oct 02, 2012 11:17 am

What do you get on pins 6/7 (they are connected anyway) of the BA6110? In theory this should be a signal about 2V peak-to-peak on a DC offset of about 5V. Same for pin 8 (so the 700 mV offset definitely is too small).

If I'm not mistaken this DC offset may come from R162 and the surrounding parts.

(Another small tidbit: Through D27 comes only a signal for accented notes; then the accent envelope is added at this point).
MMM
 
Posts: 420
Joined: Tue Apr 24, 2012 12:51 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Wed Oct 03, 2012 5:41 am

No, I dont get a AC signal with DC offset, I didnt meassure it right now, since i am at home and not at my x0xb0x, but I definetly was wondering all the time that there wasnt some kind of that signal ( AC ) at the buffer output.

The next step I did want to do, is exactly that again applying my voltage source on pin4 ( to make my ba6110 work) and then scoping the output, right now I think I will see your AC + DC offset voltage at those Pins, but tomorrow I will be more intelligent.

Thanks again.
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by MMM on Wed Oct 03, 2012 8:06 am

Please check the DC offset at pin 1 and 2 of the BA6110, too.
MMM
 
Posts: 420
Joined: Tue Apr 24, 2012 12:51 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Thu Oct 04, 2012 11:54 am

BA6110: Pin1 : 5.29 V
Pin 2: 5.29 V

This seems perfectly good, as I can tell form the schematics.

And I still know that those signals must be overlayed with my sound AC signals, because I can here the output when tricking my BA6110 into working,.

What I did test right now is the voltage needed to get my Q1-Q4 to work properly its about 670 mV although sometimes the high tunes get clipped, that voltage I did apply to the Base of Q3and Q4.
I think that the base voltage of those transistors is wrong in my case, so that the voltage applied to Pin4 is slightly too less to make the VCA work.

Another completly different point would be the Potentionemeters installes, could it be that I soldered in the wrong ones?

I had this confused perhaps, one said to be center pitched one normal, I do not really know if that matters.

At least I know I have the linear and logarithmik ones in the right place, but then it said VR2 should be center detent, I dont know if I got that one right, how can I measure it?

I got my probe at the center pin of VR2 and when I twist it the voltage rises or falls above 0V, the Pin 1 and Pin3 dont show any behaviour at all, so I guess it should be center detente, right?

Thanks and greets
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by MMM on Thu Oct 04, 2012 12:29 pm

More food for thought:

Pin 1 and 2 look indeed fine.

You said that you don't get the 5.xx V DC offset at Pin 6/7, right? But this offset is generated externally, too (if I'm reading this right). So my guess is that something's wrong with this particular bias voltage. Check along the path R162, R128 and the path R162, R129, R127.

I'm not familiar with the inner workings of the BA6110 but I could imagine that without this bias you'll get one half of the waveform at best. This could explain that you perceive some "clipping" when using your test setup.

As far as the potentiometers go: "center detent" means that the potentiometer has a point where it slightly "locks in" mechanically at the center (think of pitch bend wheels at most synthesizers); it has nothing to do with the electric characteristics.
MMM
 
Posts: 420
Joined: Tue Apr 24, 2012 12:51 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by MMM on Thu Oct 04, 2012 12:40 pm

After looking at the datasheet for the BA6110 it appears to me as if the output of the actual VCA would not carry a DC offset by itself. Since this is a symmetrical amplifier and both the positive and negative input are biased identically this will cancel itself out; and so you get just the AC signal at the output of the VCA (pin 6).
However since the chip is operated from a single supply (and not from a dual supply) the input of the buffer (pin 7) requires re-biasing. The output of the buffer (pin 8) will also carry this bias because the buffer is not a symmetrical amplifier.

I'm open for corrections if I got any of this wrong.
MMM
 
Posts: 420
Joined: Tue Apr 24, 2012 12:51 am

Re: x0xb0x newbie: Problems with VCA BA6110 pin 4

by mex on Sat Oct 06, 2012 3:37 am

Good morning again, saturday work, now it has to work :)

I just tested your suggested route, which does the dc offset:

R162 - R128: brings the correct 5,33 V where it should be and cuts it down to 0,7 Vafter R128, I guess this is right, but there could be a error at least.

The other route of course is the same with another middle voltage between R129 and R127 with around 3,3 V, but as it seems they only form another part of the voltage dividor.

Next step now I will desolder Pin 4 and apply a voltage directly to this pin, I hope then I wont cause that much current flow, but still will be able to produce some sound.

UPDATE#1

Just did desolder Q1, and I still have massive curretn flowing into Pin4 when supplying an external voltage supply about 1 V it begins to produce the sounds I play on the keyboard, after that curretn slowly rises sound gets louder but I got around 25 mA when supplying 1,7 V that cannot be right I assume.

So perhaps I have some broken BA6110 after all??? and worst case I cannot find the other VCA chip also supplied in my x0xb0x package, so I cannot even try the other one right now, that is really sad, gotta keep digging, perhaps I will find ich some day :)


UPDATE#2

Next step I will change the value of R131, I do believe I simply need more volatge t the base of Q3, so I will alter the settings of this npn transistor by changing the voltage. I will try this by lowering the value of R131 from 220k down to halt in first step simply by soldering another 220k at the bottom ( or whatever resistor I will find in the next 5 seconds :) )
Didnt change anything, I dont understand how the voltage on Q3 is built, but i definetly need a slightly bigger one to make this work.

UPDATE#3

I did attach my voltage supply to base of Q3 again, with a voltage of around 600 mV it works, so I do believe perhaps my transistor in Q31 is somehow a little broken or shaky, normally I will get a voltage on keyboard trigger fo around 500 mV, but this ain enough to pull the Q3 to work, so I reall do not need a lot but some changings in the resistor devidors, perhaps I could also tweak R123 and R132, i addition with R130 to pull something off.

So next I will try to get a higher voltage there, like i tried before, but hopefully it will work this time.

Ok i did lower the 1M5 Ohm R123 to half, no changes, I did highr R132 to 10 kOhm, no changes, When i look at R133 there is always 11,4 V when I push a key, on R131 on te upper side also, on the lower side it is different, the voltage goes immediatly up to around 6 V slopes down really fast around 30 ms down to under 1 V , when I keep the button pressed it will ramp up to 12 V again I guess, when realsing it gos down to 0 V.

ut all of this has absolutley no affect on the collector of Q31, how can this be how can this pnp transistor smoot down all of this and simply let 500 mV pass all the time the key is pushed, I do not udnerstand this, and I have no possibilty right now to alter this voltage I feel I really have to alter a little bit...


UPDATE#4

OK there comes light at the end of the tunnel, but I guess really at a long term cost.
I did solder 10 k parallel to R131 so i get around 9 k Ohm at this point, and now i am hearing sound for the first time without external power supply.

This is doing so by then pulling up the critical voltage behind Q31 to around 600 mV, sound is really not loud at the moment so I will play around with a poti or even with some fixed values later on, so that I can use my x0xx this aternoon at home :)
Last edited by mex on Sat Oct 06, 2012 5:32 am, edited 2 times in total.
mex
 
Posts: 10
Joined: Mon Oct 01, 2012 7:52 am