Latching circuit doesn't seem to work right

Thread Starter

Andy S.

Joined Mar 26, 2007
9
Folks,

Well, I came across this site by accident. I was looking for some guidance on a circuit I am trying to repair and found a similar issue here on this site. Actually, just a similar circuit. Hopefully I can get some good info here and relearn stuff I probably forgot from 20 years ago.

What I have is a guitar effects device that has a Momentary switch that "should" activate a latching circuit, turning on the distortion. Press the switch a second time and it should turn off the distortion. Situation is when the switch is pressed, nothing changes. But, I discovered that when I put a test lead on either the point of R40 or R43 where it meets with pin 2 of eihter Q16 or Q16 respectively, then the LED comes on and the circuit is active. However, it stays active ONLY if the switch is kept in a closed position.

Am I looking at a failed device in Q6 or Q4? And why doesn't the switch work to activate or deactivate the latching circuit.

Or, worse case, I am totally wrong in my thoughts on the latching circuit and there is something real obvious that I am missing.

Any guidance would be greatly appreciated.

Regards,

Andy S.


Schematic is at:

http://www.schematicheaven.com/effects/boss_bd2_bluesdriver.pdf
 

Thread Starter

Andy S.

Joined Mar 26, 2007
9
Red test lead at point near resistor mentioned and black lead, from a VOM, attached to ground. Seems like since I'm touching what is around 7.9volts and ground, I'm dumping the voltage straight to ground. Does that sound right?

I already swapped out the two transistors Q16 and Q15 as I thought eihter or both of those were the culprit. Guess not.

thoughts?

Andy S.
 

beenthere

Joined Apr 20, 2004
15,819
Q15 & 16 make a flip-flop. One should be on and the other off. The switch input should change the state. It's odd that your meter will do anything but measure the collector voltages. You might suspect a poor solder joint is causing the problem. It looks like a bad FET (Q4, 5, or 6) would not be able to mess with the flip-flop and allow an audio signal to pass.
 

Ron H

Joined Apr 14, 2005
7,063
I need to see if I understand you. Are you saying that, with the switch depressed, nothing happens until you connect your VOM to either Q15 or Q16 collector (pin 2), at which time the LED comes on? And then, if you remove the probe, the LED goes off? And the probe does nothing if you don't have the switch depressed?
I can believe that a multimeter could cause the flip-flop to toggle, due to the capacitance of the probe and VOM, or even the resistance of the VOM if you have a cheap one. But I don't understand why this would be a function of the switch position. And I don't see why the LED would go back off when you remove the probe, at least not consistently.
 

Ron H

Joined Apr 14, 2005
7,063
I was keying on the phrase it was "dumping 7.9V to ground" which made me think shorting.

hgmjr
Good point. However, as I implied above, it doesn't take much probe capacitance to cause the FF to toggle. What I don't understand is how the switch would have anything to do with that. If he wants to test the FF with his VOM, he needs to have it in Volts mode and attached to a collector, and then activate the switch. However, the LED is there as a monitor already.
 

hgmjr

Joined Jan 28, 2005
9,027
What if C30 is open or faulty? Would that give the symptom that is being described? That way if Q16 is triggered manually, then the release of the switch would turn off Q16 but with a faulty C30 the switch would not reactivate Q16.

hgmjr
 

Ron H

Joined Apr 14, 2005
7,063
What if C30 is open or faulty? Would that give the symptom that is being described? That way if Q16 is triggered manually, then the release of the switch would turn off Q16 but with a faulty C30 the switch would not reactivate Q16.

hgmjr
If C30 were open, the switch could turn off Q16 (which would turn on Q15), and it would stay in that state unless, as you say, it were triggered manually, such as by probing Q16's collector. I think that's basically what you said, except the action happens when the switch contacts are first closed (not on release).
 

hgmjr

Joined Jan 28, 2005
9,027
Situation is when the switch is pressed, nothing changes. But, I discovered that when I put a test lead on either the point of R40 or R43 where it meets with pin 2 of eihter Q16 or Q16 respectively, then the LED comes on and the circuit is active. However, it stays active ONLY if the switch is kept in a closed position
Unless I missed a later clarification, this is the excerpt that lead me to believe that the switch was in the closed position when the lead was touched to the collector and turned on the led. It then stayed on until the switch was released.

Have I got the N.O. and N.C. swapped in my head?

hgmjr
 

Thread Starter

Andy S.

Joined Mar 26, 2007
9
Unless I missed a later clarification, this is the excerpt that lead me to believe that the switch was in the closed position when the lead was touched to the collector and turned on the led. It then stayed on until the switch was released.

Have I got the N.O. and N.C. swapped in my head?

hgmjr
Ok, sorry, let me try to clarify further.

Andy S. said:
then the LED comes on and the circuit is active. However, it stays active ONLY if the switch is kept in a closed position
The circuit I mention here is the actual distortion circuit, not the FF. And the Momentary open switch is pushed closed, the distortion works. When I release the switch (open), the distortion stops working, but the LED stays lit.

I was briefly thinking about C30 but then let it go. But now...maybe that's it?

Hope that clears it up. let me know if not.

Andy S.
 

Ron H

Joined Apr 14, 2005
7,063
Ok, sorry, let me try to clarify further.


The circuit I mention here is the actual distortion circuit, not the FF. And the Momentary open switch is pushed closed, the distortion works. When I release the switch (open), the distortion stops working, but the LED stays lit.

I was briefly thinking about C30 but then let it go. But now...maybe that's it?

Hope that clears it up. let me know if not.

Andy S.
I can't see a failure mode that could cause that behavior. Has the unit been modified?
Let's take this a step at a time. Can you cause the LED to toggle on and off with alternate pushes of the button?
 

Thread Starter

Andy S.

Joined Mar 26, 2007
9
I can't see a failure mode that could cause that behavior. Has the unit been modified?
Let's take this a step at a time. Can you cause the LED to toggle on and off with alternate pushes of the button?
Ron,

Has the unit been modified.
Sort of, but not in regard to the transistors that operate the latching function. It is primarily the caps in the processed signal line that were changed. So the transistors and caps for the circuitry that turn on and off the distortion circuit are the original parts.

Can I cause the LED to toggle on & off with alternate pushes of the button?
No. The only way the LED comes on is if I touch the red probe to the aforementioned points. And, the meter is in the Voltage mode, and it is a Fluke 8060 Multimeter, since that was questioned a bit earlier.


Ever since you mentioned C30 being open and locking the FF, that has been rattling around in my head. That sounds pretty much like what is happening. Since it (C30) is open, it prevents the FF from behaving normally and responding to the momentary switch. Thus, once I trigger it manually (with the probe) it activates a change in state, activating the change in the processed signal path.

But, since the cap may be open, the circuit doesn't stay in that position, it reverts back to its original state. The circuit does stay ACTIVE as long as the switch is pushed down and contact is maintained. Does that sound feasible?

If I get a chance this afternoon, I'm going to check that cap, or just swap it out and see what happens.

(MAN!!! If only I'd paid a little more attention in basic circuits class 22 years ago!!)
 

hgmjr

Joined Jan 28, 2005
9,027
A literal interpretation of the schematic would indicate the switch pictured is normally closed. That is, when the momentary switch is not pressed; the control line is grounded.

That makes me think that when it is pressed the switch contacts are not connected together. It is important to the troubleshooting process to establish with certainty whether the closure to ground occurs when the switch button is pressed or is the closure to ground provided when the switch is released.

You may have already answered this question earlier but just in case not, is this a system that was working and failed or did you acquire this system only to discover that the on/off control was not working?

Are you certain that the momentary switch that you are using is providing this normally closed condition when the switch is not pressed? If this is not the case, then the operation of the FLIP/FLOP may be confused and behave in a manner similar to what you have observed.

hgmjr
 

Ron H

Joined Apr 14, 2005
7,063
A literal interpretation of the schematic would indicate the switch pictured is normally closed. That is, when the momentary switch is not pressed; the control line is grounded.

That makes me think that when it is pressed the switch contacts are not connected together. It is important to the troubleshooting process to establish with certainty whether the closure to ground occurs when the switch button is pressed or is the closure to ground provided when the switch is released.

You may have already answered this question earlier but just in case not, is this a system that was working and failed or did you acquire this system only to discover that the on/off control was not working?

Are you certain that the momentary switch that you are using is providing this normally closed condition when the switch is not pressed? If this is not the case, then the operation of the FLIP/FLOP may be confused and behave in a manner similar to what you have observed.

hgmjr
I agree that this should be checked. The FF, when working correctly, will only toggle when the switch contacts close. I think you are seeing artifacts of several copy generations. See below.
It wouldn't seem to be desirable to have the FF toggle when you release it.
I still can't see a failure mechanism that would explain what Andy describes. If the LED is on, then the audio bypass FET will be off.
 

Attachments

hgmjr

Joined Jan 28, 2005
9,027
A flaw in my assumption negated my explanation so I defer to RonH's explanation in the post that follows.

hgmjr
 

Ron H

Joined Apr 14, 2005
7,063
Subject to validation by RonH, here is my brief theory of operation for the flip-flip circuit in question.

Begin by assuming that the FF is in the state corresponding to Q16 being in the ON state (the collector at or near 0 volts) and Q15 in the OFF state (the collector at of near the positive power supply).

The control switch is not pressed so that the control line is at ground.

The above assumed conditions have been present for several seconds at least so that the circuit has reached a steady-state.

At this point C30 has charged to around 0.7 volts (Vbe) and C32 has completely discharged so that little or no voltage is present across the capacitor.
Now assume that the switch is pressed and the contact to ground is opened up as a result. Current is delivered to the common connection between C30 and C32 from the positive power supply voltage through R45. Since the voltage across the capacitors cannot change instantaneously then the resulting current will be routed to the base of Q16 through C32. This is because the voltage across C32 is nearly zero while the voltage across C30 is around 0.7 volts. This current pulse applied to Q15's base is sufficient to turn on Q15. As Q15 turns on, the transition of its collector drives the base of Q16 low through R44 and the capacitor C28? that is in parallel with R44. This turns off Q16.

At this point, the LED should be lit and the switch can now be released to re-arm the FF for the next switch depression to change the FF back to the originally assumed state. The transition back to the original assumed state is pretty much identical with the exception that it involves the components that correspond to the ones already mentioned that are used in the other half of the FF circuit.

hgmjr
Well, first of all, you have your C30 and C32 voltages swapped, but you are barking up the wrong tree. Nothing happens when the switch opens.
With those initial conditions, it works as follows:
The base of Q15 is at 0V, the base of Q16 is at 0.7V (opposite of what you were thinking). When the switch is opened, C33 (0.01uF) charges very slowly through the 1Meg resistor. Very little current will flow through C30 and C32 (I=C*dv/dt). Almost all the current from the 1Meg goes into the 0.01uF cap. Even if this cap were missing, and all the current flowed through one of the 220pF caps, the 9uA from the 1Meg would not be enough to forward-bias the OFF transistor's base.
However, once the 0.01uF cap is charged, then when the switch closes to GND, the 0.01uF cap will discharge very rapidly through the 100 ohm resistor, and this transition will couple through both of the 220pF caps. This will have no effect on the transistor that is already off, but it will reverse bias the base of one that is on, causing its collector to start to rise, which begins to turn on the other transistor. This transistor's collector starts to fall rapidly, reinforcing the turnoff of the 1st transistor, and the toggling is completed.
As I said, in the schematic, you are seeing artifacts of multiple copies that makes the switch look normally closed. The toggle action occurs when the switch closes. Nothing is supposed to happen when it opens.
I still don't see how the LED can be ON while the non-distorted path is enabled, and the stated FF symptoms also occur, unless something is wrong with one or more of the JFETs (or their biasing components) AND something is wrong with one or more of the FF parts.
BTW, I simulated the FF, and it works as I described. The simulation was to see if I could induce the described faults, but then it dawned on me that something else was wrong (as described above), or Andy is inadvertently not giving us an accurate description of the symptoms.
 
Top