BA662 vs BA6110 thread, again

[Dimitree]

You have mentioned "if pins 7 and 8 are not used, is pin 4 used?  id guess not"
but in the CS-2 circuit the buffer is used and pin 4 is biased from the R10, R12, R9 divider.

as I said before, the CS2 circuit shows it as used..but if you look inside any CS2 (or search for internal photos on the web) you will notice that 7 and 8 are not connected.

> the ba6110 has the 1.2V input voltage drop

As does the LM3700.

The '3080 Iabc input is simple, the '3700 Iabc input is more precise at extremes (which may not matter here, as long as we work to the higher voltage).

There's really not much else can be different.

maybe I'm not understanding, so if there are no differences, why should I choose a LM13700 instead of BA6110, if in both options I need to adapt the chip to the circuit?

the ba6110 buffer is identical to the ba662 buffer, which is why its a good substitute for a lot of applications.  the only difference is that the ba6110 doesnt have the bias control pin, which shouldnt matter too much.  if it does cause issues, a discrete solution is the only other way.

true, but I missed 1 thing: pin 3 of BA6110 is "input bias", how should this pin be connected? there's no such pin on the BA662

[armdnrdy]

I've been working on the same issue concurrently. BA662 to BA6110 conversion.

This has already been achieved with the x0xb0x using a current mirror.
BYOC designed a Boss VB-2 using the same conversion configuration. I've reversed images of the circuit board to confirm this.

The x0xb0x connects pin 3 of the 6110 to ground. The VB-2 leaves pin 3 unconnected.

Searching the synth sites, I have read many posts stating that there is no audible difference between the BA622 and the 6110 with a current mirror on the control pin. (The current mirrors "fix" the BA6110 control pin threshold issue) (I wouldn't worry too much about matching the current mirrors in a compressor circuit.)

I read a post stating that after the comparison, the x0xb0x builder left in the 6110 and saved the 662 for a spare.

I took a look at the CS2 board images on the net. The board does differ from the schematic. Here's my take on what I saw.

Pins 7 and 8 are not connected......neither is pin 4 per se.

Since pin 7 and 8 are clearly not connected, then connecting something to pin 4 would have no effect whatsoever.

The trace that "connects" pin 4 is merely a path through pin 4 to connect R11 (150K) to bias. (R12)

So, even though pin 4 is connected.....it is not in use.

This is the configuration of the x0xb0x and the VB-2:

[g_u_e_s_t]

pin3 on the ba6110 controls a bias circuit to linearize the input stage.  either leaving it floating or connecting it to ground will turn it off, and make it the same as the ba662.  i generally like to connec things to ground, just to ensure against stray flux leakage or hum pickup.

also, on the above schematic, the base and collector of q3 should be shorted.

i spent today laying out a ba662 clone.  it fits in a 9-pin sip package which is just slightly taller than the original.  hopefully it fits inside peoples boxes.  also, hopefully it works!

[armdnrdy]

Good catch! It was hastily drawn on my way out the door.

Revised and reposted.

[armdnrdy]

i spent today laying out a ba662 clone.  it fits in a 9-pin sip package which is just slightly taller than the original.  hopefully it fits inside peoples boxes.  also, hopefully it works!

Is this made up of discrete components? Or a 6110 with current mirrors?

[g_u_e_s_t]

Is this made up of discrete components? Or a 6110 with current mirrors?

it will be discrete components.

[Dimitree]

update after some experiments:

I did a in-depth comparison between BA662, BA6110 (with no current mirror) and CA3080E, using the original CS2 and a breadboard, I desoldered the chip and soldered some cables in there, so I could change the chips using the original pedal.
I then recorded the 3 chips, to do that I recorded some guitar riff so there were no playing differences.

This time BA6110 had pin 3 connected to ground, and now it works, and (again) I didn't use any current mirror (I don't have PNPs here)

The 3 chips sound almost identical, only difference I noticed is that 3080E is a lot noisier, it even catches some radio frequencies (I clearly heard a DJ speaking while I was not playing and my hands were near the chip!!). It has more than 6dB of noise than BA6110 and 9dB of noise than BA662.
The output level is almost the same..
I will do other experiments (I only tried with Sustain at max and Attack at max) and analyze these 3 clips again in different domains.
Perhaps I should try with hot signals to test that 1.2V difference?

[Dimitree]

I made this quick animated image
don't know if this is useful to undetstand differences...but for sure it shows different noise floors:

[g_u_e_s_t]

the extra 3080 noise is not suprising, those are old chips using an old process.  its interesting that the ba662 is less noisey than the ba6110.  perhaps there is a bit of noise added from the extra transistors inside for doing the linearization.

the 1.2v problem will show up at low input signals, with low sustain.  this is when the ba6110 will have the smallest voltage at its control pin.  for larger signals, the relative difference of .6V wont be as big of a deal.

[wookie]

Resurrecting an old but not ancient thread here....

I ended up stumbling into this issue myself by way of the BYOC 5 knob compressor (see relevant thread at:  http://www.diystompboxes.com/smfforum/index.php?topic=106649.0).  I see some mentioning of grounding pin 3 (input bias current) of the BA6110, but has anyone tried feeding it 200 uA?  That seems to be the suggestion of the datasheet.  If the BA6111 is hooked up with Vcc at 9V and Vee at 0V, a 47k resistor between pin 3 and Vcc should get you really close.

That said, the BYOC 5 knob compressor leaves pin 3 unconnected.  I found the Ibanez CP-10 uses the BA6110 does the same.

BA6110 Datasheet I'm referencing is here:
http://urekarm.tripod.com/synth/BA6110.pdf

[Yazoo]

I've just built a CS-2 using the BA6110 chip. I used the Boss pcb as a model and removed the FET switching, using true bypass instead. This left room on the pcb to put the current mirror in. It was then a case of moving components around to match the BA6110 pinout. I used the output buffer, pins 7 and 8. I did see from the PCB photo I found that these were left disconnected but I'd found another version where they were left in and I decided to go with that.  I grounded pin 3. I'm pleased to say it's all working very nicely.

[armdnrdy]

I too have used the BA6110 as a replacement for the BA662 in a build. My connections were as in the "conversion" schematic dated 12-23-14 that I posted.

[Yazoo]

Thanks, I used your diagram and found it very useful. I just had to rejig the tracks on the Boss pcb to re-route the connections to pins 1 and 4 of the BA6110. Once I had taken out the FET switching, this became a lot easier.

[armdnrdy]

Good..I'm glad that it was useful.

[kbibs]

I have an extra lm13700 and was thinking of building a CS-2 clone with it. Just curious, has anyone successfully done this? ... If so, what did you think of it?

Thanks!

[armdnrdy]

The Dr J D55 Aerolite Compressor is basically a clone of the CS-2 with some other bells and whistles.

It uses a 13700 instead of the BA662. Look to this schematic to see the 662/13700 conversion.

http://s247.photobucket.com/user/mysticwhiskey/media/Joyo%20DrJ%20D55%20Aerolite%20Compressor/DrJD55AeroliteCompressorSchematicV01_zpsddb46117.png.html

[kbibs]

armdnrdy,
Thanks for the link to the D55! ... I didn't realize it was a clone of the CS-2. I'll definitely take a close look at this one. It's funny cause I was also thinking of adding a blend knob also and the D55 has the mix knob. It looks like it uses a mu-amp to boost the original input to match the wet signal where I was thinking of using an op-amp.

I'm also going to refer to the .pdf file that I downloaded from this site http://www.eserviceinfo.com/fulltext.php?search_ft=LM13700.

I'm glad to see this has been done before.

Thanks Again!

[kbibs]

Happy New Year All!,

I just breadboarded a Boss CS-2 using an LM13700 in place of the BA662 and following the marked up Boss schematic in the link I posted above. The only issued I had with the mark ups was biasing the linearizing diodes up to 9vdc through the 15K. When I connected that the compressor had a very low output, less than unity at low compression settings. Not connecting, i.e. open bias, the compressor had above unity even for the min compression(sustain) setting. I don't quite understand this but with it open it sounds very very good. Good enough to box up!

If anyone could explain the low output with the linearizing diodes biased up I'd really appreciate it.

Thanks,
Ken

[armdnrdy]

Happy New Year All!,

I just breadboarded a Boss CS-2 using an LM13700 in place of the BA662 and following the marked up Boss schematic in the link I posted above. The only issued I had with the mark ups was biasing the linearizing diodes up to 9vdc through the 15K. When I connected that the compressor had a very low output, less than unity at low compression settings. Not connecting, i.e. open bias, the compressor had above unity even for the min compression(sustain) setting. I don't quite understand this but with it open it sounds very very good. Good enough to box up!

If anyone could explain the low output with the linearizing diodes biased up I'd really appreciate it.

Thanks,
Ken

Did you notice that the schematic of theD55 Aerolite Compressor that I posted doesn't connect the linearizing diodes?

http://s247.photobucket.com/user/mysticwhiskey/media/Joyo%20DrJ%20D55%20Aerolite%20Compressor/DrJD55AeroliteCompressorSchematicV01_zpsddb46117.png.html

[kbibs]

armdnrdy,
Yes I did, Thanks! ... that's what gave me the idea to leave it open. The app notes for the LM13700 show it connected with a 13K resistor for a voltage controlled amplifier application but the front end circuit is different than the CS-2. The front end to the D55 is also a little different than the CS-2, it's more like a Dyna/Ross.

Ken