Help - This gator is killing me and my 13600s

[blackieNYC]

I would really like to have a @&#$! gator. I have found substitute info for using an LM13600 in place of the CA3080, but I am convinced that the LM13600 cannot handle having its ground pin manipulated the way that the 3080 does. They pass audio, and then get hot.  And then they are probably shot - low max current.  The pin sub I use is
3080       13600
2   -in          4
3   +in         3
6   out         5
5  bias/gain  1
4 V-/ctl/gnd  6
7  +9           11
unused stuff - I have the 13600 buffer ins and outs grounded (pins 7&8,9&10)and the out and in of the unused half are shorted together (12&13) and pin16 (unused bias/gain pin) is grounded - these are as per Merlin's Eng Thumb schematic.
     Shouldn't this be done with the 13600 pin 1?  [edit: I'm back on this after giving up some time ago. forgot the earlier thread.  RG suggested the use of a current mirror, Kipper has a drawing, but I don't know how to implement this.
http://www.diystompboxes.com/smfforum/index.php?topic=117220.msg1088040#msg1088040]  Should pin 1 simply get the control signal right from pin 7, and pin 6 gets grounded? and the gain control becomes a voltage divider after the output? That would look a bit like the Tri-vibe's LFO input. What do you suppose would have to change?   
I don't know what's up with gaussmarkov.net but here's a schematic from an older forum post:


[ edit: So do I make the control voltage the base& collector primary side, and the mirror side feeds pin 6?  Pin1?  Can those darlington buffers be utilized or does that gain pose a problem?]

[armdnrdy]

If you have any doubts as to the similarity of the 3080/13600....read this....straight from the horses...err...the designers mouth.

http://www.idea2ic.com/LM13600/LM13700.html

If the 3080 works in the circuit...the 13600 will as well...as long as you are terminating the unused portion properly.

[duck_arse]

the problem might stem from the unused parts being grounded, while the used part is not grounded. perhaps tie the unused parts grounds to the used parts controlled ground [?or to nothing?] and see if the parts survive.

[armdnrdy]

Here is the correct way to terminate the unused section of a LM13600/700.

If I recall correctly....the question was answered by an Texas Instruments engineer on a T.I. blog.

https://www.dropbox.com/s/mg9o99rhzosd8em/LM13700%20-%20CA3080%20adapter%20board.pdf?dl=0

This configuration puts the unused section in a "known" state.

[stringsthings]

Hey! That's my old schematic I wrote up way back when.

I'm old enough to remember the "good-old days" of PAIA way before the interwebz made pedal DIY much more viable.
It was pretty hit or miss back in the day.  But I digress. 

Tagboard has another vero layout with a 13700.
http://tagboardeffects.blogspot.com/2016/10/paia-5730-gator.html

[reddesert]

My guess is that there is a mistake somewhere that is causing the chip to conduct or putting voltage where it shouldn't be.

Honestly, I would try it without the proper termination, just with the pin-for-pin substitution for the 3080. If that works, then you've eliminated one possible source of the problem and can go back and try to get the termination right. I have built just one circuit this way with a 13700 and it worked fine, which doesn't prove that a Gator will work, but it is highly likely that it can work.

[armdnrdy]

My guess is that there is a mistake somewhere that is causing the chip to conduct or putting voltage where it shouldn't be.

Honestly, I would try it without the proper termination, just with the pin-for-pin substitution for the 3080. If that works, then you've eliminated one possible source of the problem and can go back and try to get the termination right. I have built just one circuit this way with a 13700 and it worked fine, which doesn't prove that a Gator will work, but it is highly likely that it can work.

If you read the OPs first post..."unused stuff - I have the 13600 buffer ins and outs grounded (pins 7&8,9&10)and the out and in of the unused half are shorted together (12&13) and pin16 (unused bias/gain pin) is grounded"....
This isn't really the "proper" way to terminate the unused half of the OTA.
This may work in Merlin's Eng Thumb design but....that is a different circuit.
I would suggest trying the voltage follower configuration depicted in the PDF that I posted.

[blackieNYC]

Thanks guys. That tagboard layout does do what I have on the breadboard, or so I think. That is, the ground pin is manipulated by the control signal and the bias input of the 13600 is given a gain control pot to pin 1. I'm sure the tagboard is confirmed, but other than that, I have not seen (nor heard of here I think)any schematics (including those informative links thx, Armd) that connect the ground to the control signal of a 13600 -it's always connected to pin 1. This, and the fact mine doesn't work led me to believe the gator with a 13600 needs to be rethought.  I don't see how the unused pin terminations from the ET would not be right, but I'll try these recommendations.  PRR told me the 13600 should work by varying the connection the the ground pin, or the bias pin.
Better order more 13600 from Small Bear, and tear up the breadboard again.

[blackieNYC]

Using the 13600 in place of the 3080 in the gator is... challenging. This application appears to NOT be a simple pin-for-pin conversion. It's pin 2.  I've got it sort of working in an awful manner.  I built a little data sheet test circuit and made sure my 13600s were good.

First of all, if I don't give the 13600 pin 2 diode bias a voltage, there is no output at all.  I don't have it working fine, but I believe a 13K resistor is necessary from 9v to pin 2 only because the ground pin 6 is the one being varied.  Yes, as PRR said, it should work, but there is no such use of the V- or ground pin in the data sheets or in any 13600 pedal application I can find. This method seems stupid unique to the gator. (just using the 1-8 half of the 13600.  The unused side is 4.5 v to pin 14, 12 & 13 jumpered)

What should that varying voltage look like?
At IC3C pin 4 I have 9v when there is no tone, and 4.5v when there is tone (100mv p-p).  At pin 6 of the 13600 I have 4.5v with no tone, and this drops slowly (looks right) when tone is applied to 1.5V. 
Signal - the output of the IC1is fine (unity), but pin 5 of the 13600 yields about 20mv, and IC3a pin 12 has the signal at 50mv p-p.  So the output is below unity. But it's not working quite right.
The sound (guitar)- the note fades in slowly (seems ok, but at this point it is weak and very distorted), then it seems to drop down for some milliseconds, followed by a somewhat louder clearer version of the note. This second surge of the note is much like at echo - a second note entering as the first is decaying.
And, of course, the finale is an excessive thump as the gate snaps shut.  & I must have the threshold and gain maxed to get anything.

[blackieNYC]

Bit of a bump. If the Lm13600 gain cannot be manipulated via the ground pin in the same way as the ca3080, can this control signal simply be connected to pin 1? As in most modulators? Would would it take?
- can it deliver sufficient current?
- voltage is 4.5v or 1.5v, off or on
Would this do for turning on and off via pin 1?  Would zero v be neccessary?
And

What would control gain? In the gator pin 1 is used to dial up a fixed gain (fixed meaning at whatever you dial in).  See the runoffgroove tri-vibe. Pin 1 is the on/off sweeping modulator. What is determining the maximum output level to be unity gain?

Armdnrdy has this fencepost gate at madbean. I think if this can be changed to gator-like operation this would be a simpler gator. Maybe figuring out how to get the gator's swell control line into the fencepost would get me there
http://www.madbeanpedals.com/EP/schematics/Fencepost.gif

[idiot savant]

If you can't get it working by terminating the unused half of the 13700, then why not try paralleling both sides? You should supposedly be able to get a marginal improvement in noise that way as well.

Using the V- pin for the control voltage is something I haven't seen outside of the Gator, but there are some other schemes that use it in slightly different ways. Check out a few of these links:


discussion about the Jung current source, including a little talk on paralleling:
http://electro-music.com/forum/post-347005.html

a little more on the Jung trick:
http://electronotes.netfirms.com/AN399.pdf