opamp weirdness

[markr04]

I built a Shaka Tube from Paul's layout twice. Both times I have the same problem: a loud strum causes a fartgate sound and the output goes dead. But I found something weird: when I touch my DMM's positive probe to the input on the opamp, the problem disappears and it sounds great. The negative probe is clipped to the "ground" of the AC wallwart. My DMM is set to measure voltage... the trick doesn't work when set to measure resistance.

I can touch any solder spot within the red box I drew and clear up the problem:


I'd rather not have to find a bigger enclosure to fit the DMM in with the effect if at all possible.  :wink:

[ErikMiller]

What is the color code on the 470K resistor?

[markr04]

Yellow, Purple, Yellow, Gold

Maybe lowering that value a bit would do it?

[markr04]

I forgot to mention:

1) After the output fartgates to zero and I leave my guitar alone, the output will return in about 1 minute... I hear the low noise and hum return.

2) The gain pot has no effect unless I touch the DMM probe to the circuit as I described earlier.

[Peter Snowberg]

It sounds like a bias problem.

What are the voltages between:

A) U1 pin 7 & U1 pin 4

B) U1 pin 4 & U1 pin 3

C) V- & In Gnd

That should point to the problem.

[ErikMiller]

I'm thinking it's a part value somewhere around the opamp. Verify all the part values and orientations around the opamp. The .01 cap, the 250pF cap, the 2.2uF cap, the 220 ohm resistor., the 470K resistor (you already verified that).

It's a good job for the troubleshooting combo of signal generator and oscilloscope (or audio probe).

You're using a proven schematic on a proven layout, so all that's left is incorrect parts/orientation. Got the 2.2uF pointed the right way?[/code]

[markr04]

Please look at what happens to the voltage at pin 2.

Voltages at U1 (when it's working):

1: -10.6
2. 7.8
4: -10.6
5: -10.6
7: 13.4
8: 0

At U1 after it's fartgated to zero:

1: (this "resets" it the same as touching pin 3, so -10.6)
2: -9.4** (this is interesting!)
4: -10.6
5. -10.6
7. 13.4

I lack the experience to know what those point to.

[Peter Snowberg]

Please post the voltages between the points in my post above and that will point the way. In measuring, it doesn't matter which probe goes to U1 pin 4 and pin 7, just that those are the two points to measure from.

A set before and after "fart out" would be helpful.

[BJF]

Hi,

If you measure voltage with a DVM, you can assume you have a 10M Ohm resistor between the probe ends; your finger has a resistans to ground
( through your body) eqvivalent to about 10M Ohms; this is a JFET OP and it needs only a few pA in biascurrent; pin 3 is reference input and will set voltage at pin 2....if pin 3 does not have a reference voltage OP amp will be confused
You have created a path to ground while measuring or probing with your finger......
Try connecting a 10M Ohm resistor from the red block to ground what happens  and what does that tell you?

There can be a small crack in the pcb trace that is shorted when you probe.

Just a thought
BJ

[puretube]

I used the search function to search for a shaka tube schemo, Paul, and (complete) layout......

have you measured the OK-ness of the gain-pot (including the connections) yet?

[casey]

I built a Shaka Tube from Paul's layout twice. Both times I have the same problem: a loud strum causes a fartgate sound and the output goes dead. But I found something weird: when I touch my DMM's positive probe to the input on the opamp, the problem disappears and it sounds great. The negative probe is clipped to the "ground" of the AC wallwart. My DMM is set to measure voltage... the trick doesn't work when set to measure resistance.

I can touch any solder spot within the red box I drew and clear up the problem:


I'd rather not have to find a bigger enclosure to fit the DMM in with the effect if at all possible.  :wink:

all you have to do then is to get a very large enclosure and solder your meter's probe to the pin and box the whole thing up.....just run the voltage to your meter before you box it to always make sure that it can stay on.    :roll:

[markr04]

Peter:

7&4: error (DMM shows -1, which usually means error I think)
4&3: -1.1v
V- & GND in: -1.5v and falling. Measuring between these two points causes the opamp to go quiet.

[Paul Marossy]

I etched my PCB per the layout I have on my Shaka Tube page, but when I built mine, I actually modified it for use with 1/2 of a TL072. It seems that everyone who has tried to build one from my layout using a single opamp has had this gating problem. The evidence seems to suggest that using a single opamp here instead of the dual opamp originally specified causes some sort of biasing problem with the opamp.

Does that make sense to anyone? It sure doesn't make any sense to me.  :?

[markr04]

Paul,

I'm heading out to my local supplier to get some TL072s. I will try your suggestion when I get back. Will you please post what will need to be connected from the layout to make the 072 work? Maybe like:

socket pin 1 -> 072 pin 3
socket pin 4 -> 072 pin 2
etc?

[puretube]

Peter:

7&4: error...

BINGO!     (oughtta be ~25V, IIRC  :?: )


[edit]: PT goes to the search button again, to look up schemo, layout, etc..., to explain, that there might be a PS centrepoint not grounded, or similar..., or reversed PS diode or whatnot...

schemo
PCB/layout


...or a layout-error?

[Paul Marossy]

Paul,

I'm heading out to my local supplier to get some TL072s. I will try your suggestion when I get back. Will you please post what will need to be connected from the layout to make the 072 work? Maybe like:

socket pin 1 -> 072 pin 3
socket pin 4 -> 072 pin 2
etc?

Sure-

For a dual opamp conversion, connect the pins as follows:

Pin 1 - Take connection to Pin 6 and change it to Pin 1
Pin 2 - no change required
Pin 3 - no change required
Pin 4 - no change required
Pin 5 - I don't remember what I did with it. Connected to ground?
Pin 6 - connect to Pin 7
Pin 7 - connect to Pin 6
Pin 8 - Take connection from Pin 7 and change it to Pin 8

Dual opamp:

Single opamp:

[puretube]

Peter:

7&4: error...

BINGO!     (oughtta be ~25V, IIRC  :?: )


[edit]: PT goes to the search button again, to look up schemo, layout, etc..., to explain, that there might be a PS centrepoint not grounded, or similar..., or reversed PS diode or whatnot...

schemo
PCB/layout


...or a layout-error?

:?:

[Paul Marossy]

markr04-

I don't know if this will help you or not, but here is a picture of my modifed PCB



I disconnected all of the connections to the foil at Pin 6 and jumpered the output with those two black jumper wires. Pins 6 & 7 are connected together. Pin 5 is floating (maybe I should do something with that). Pin 8 has a small jumper to V+.

In essence, I cut three foils, added three jumper wires and connected two of the opamp pins together. It's really pretty simple to do on my layout as it is currently.

Well, now I need to eat my lunch (I'm at home on my lunch hour writing this!  :wink: )

[markr04]

Please eat your lunch, Paul!

Thanks for the pic, as that's my next step.

BTW: I played with the 470K, going to 510K, 390K, 150K and it didn't help.

[markr04]

I just tried the TL072 and your mod to connect it properly. It behaves in the same manner.