Build Your Own Clone Message Board

I just built this unit today and wired it all up. Everything works, but the bypass switch doesn't disable the effect.
When I press Bypass, the LED toggles on/off correctly but the delays are generated on new inputs in both switch states. Everything else seems to work great... tap tempo works, mod depth and rate work... dry and mix output both have a signal... repeats can make the thing do crazy infinite oscillations.

Any idea where I should be looking for maybe a faulty solder joint? The switch seems to be in good working order since it drives the LED correctly, I just can't bypass the effect at all.

H

What does bypass do in trails mode?

The bypass LED working correctly doesn't necessarily mean the MCU is outputting voltages to the bypass JFETs. In normal mode bypass, you should have 5v at point C (pin 7 of the PIC16F684) and 0v at B (pin3) and 0V at A (pin5). In trails mode bypass you should have 5v at B and C but 0v at A. When the effect is on (regardless of mode) you should have 5v at A and B, but 0v at C.

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

I have 5V at pin 3 and 5 and 0V at pin 7 whether I turn trails on or off. With the switch off, I have 0v at pin 3 and 5 and 5V on pin 7.

I traced through the board from pin 7 towards the JFET. The voltage is there with the switch off but it hits the 1N4148 diode and the voltage is cut off. From the instructions on page 9, it says to insert the diodes with the stripe side in the square solder pad matching the PCB. I did that, but that means current flow would be blocked from the MCU to toggle the JFET via the gate pin.

I think the instructions are wrong and the A B and C diodes (4) are installed backwards.

If your theory is correct, then neither the A or B switching FETs should be in the open position, so you wouldn't be able to hear any echo at all, and none of the Echo Royals that came before yours should work correctly.

It seems more like your switching FETs are all stuck in the open position. Which model FETs were in your kit? Are you able to use the MIX knob and adjust the output so that it is 100% wet delay signal or is there always a dry signal mixed in?

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

I agree it doesn’t make sense as others would not work before me.
I got J111 which are drain, source, gate left to right with flat side forward.
The diodes are coming off of the gate pin on the J111 and head towards the microprocessor.

Maybe I have to try to troubleshoot the schematic again but I thought the voltage coming from the microprocessor carried the voltage to use on the gate pin which is being blocked by the diode.
Is the JFET supposed to be feeding voltage back to the processor from the bottom pin (looking at my pic)?

The blend knob? I’ll have to retest, but I think I had it at 100% and it wasn’t full wet (kills the initial input and only output the delay repeats). I think I always heard the dry signal.

Knowing that they are J111, do the pins make sense in the orientation on the board and in my pic?
Are there any specific voltage test points I should try or troubleshooting tests that would help?
I’m pretty sure the drain and source pins on all 4 JFETs were at 5V with the switch on or off. Ie always stuck open. My assumption was that the switch sent voltage to the processor at the bottom middle which sent a signal out on other pins (eg 3/5/7) to the gate pins of the J111 which would toggle them on/off. When I press the foot switch, I can see voltage changes from the processor all the way to one pin of the diode but then the other side of the diode is 0V because the diode is killing the flow (backwards). Maybe I’m not understanding the circuit flow though. I am a n00b here.

To visually think, I’ve attached part of the schematic that DOES seem to support the diode blocking the current. ie schematic matches instructions.

The GREEN arrow is the foot switch passing a signal into pin 4 of the MCU via the BLUE line.
From there, the MCU can use the 5V in pin 1 to optionally send voltage down the line through pin 3 to diagram point B on the RED line. That continues above at the other point B RED line.
The signal then goes into a diode circled in ORANGE which is directing signal down away from the JFET at the YELLOW arrow. If this is by design, I don’t understand what is happening here

If that is true, I likely don’t understand something about the flow of this circuit correctly which is the most likely scenario.

I need some help figuring out some next steps and things to try. So far I’m learning a lot and getting closer.

Wait. Again, I think I'm showing my n00bness.
From what I NOW understand, the flow of electricity is intended to flow from the drain to the source. When the gate becomes "more negative" than the source, the flow will redirect through the gate or in other words block off access to the source.

So, maybe that's why the diode direction makes sense... because the MCU is changing the voltage on pin 3 which extends to the gate pin, but the diode forces the direction of flow back to the MCU away from the source effectively turning off the remaining circuit.

???

I re-tested some sample points on the board... both sides of the diode when the footswitch was on and off. The values ARE changing. The first time I tested this, they were "close enough to zero" that I didn't see that the values were actually changing.
I actually tested all 4 diodes in the bottom right hand corner of the board. The other pairing had the exact same results. I'm assuming one set is for bypass and the other is for trails?

The MCU side of the diode switches between 0v and 3v while the JFET side of the diode switches between -0.05v and 0.01v respectively.
What should the readings be?
Are these voltage values too small to force the J111 to close?
Is that related to a voltage drop from the diodes or are the wrong diodes being used?
Maybe the J111 are not compatible for this use in this configuration? The datasheet says that the gate-source cut-off voltage is 3-10v. I'm only getting -0.05v on the transistor side of the diode.

Let's just focus on one thing at a time, please. This is important in understanding what your pedal is doing or not doing. Please plug your pedal in with the mix output going to your amp and test the mix knob. When you turn the mix knob clockwise, does it cut the dry signal to the point where there is 0% dry signal at full turn clockwise, or does it only control the volume of the wet signal and the dry signal is at a constant level?

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

100%. I'm with you. I'm just trying to learn and understand.

I just tested. Guitar into pedal, mix output to amp; solo pedal by itself.
The blend dial only controls the output of the repeats. I always hear the dry signal. It behaves the same with the bypass switch both on and off. The blend dial does change the volume of the wet signal; all the way CCW is 0% wet volume and all the way CW is 100% wet volume... but always with a mix of 100% dry signal.

This means that all of your switching FETs are stuck in the "open" position.

When the pedal is engaged, only the two A JFETs and the B JFET should be open, allowing the wet and dry to mix at the IC1a output buffer. If you turn the mix pot full turn clockwise, this cancels out the dry signal that passes through C20. But it sounds like the constant dry bypass signal from C14 is being allowed to mix at IC1a as well.

So it may sound like your Echo Royal is "working" when engaged, but it's not working correctly. The mix knob is supposed to be a wet/dry blend like you would find on an AD-80, not a delay level with a static dry signal like you would find on a DM-2.

Your voltage readings suggest that the MCU is working just fine, so I would assume that all 4 of the J111 are bad. You can test this by removing 3 of the J111 (leave the second A JFET located in the the lower right corner as it only needs to be closed when in trails mode) and then using jumpers. Jumper the source and drains together of the A and B for "on". Remove those jumpers and jumper the source and drain of C for bypass.

If this turns out to be the case, please email sales@buildyourownclone.com and we'll send you a new set of FETs.

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

So, desolder and remove the 3 JFETs (A, B, and C) indicated with yellow letters and jumper the top and middle holes together on A and B (shown as green lines). Leave C with nothing.
That will simulate the pedal in “on” mode.

Remove the jumpers from A and B (green lines) so there’s nothing connected then add a jumper on point C on the top and middle connections (shown as red line).
That will simulate bypass mode.

I don’t have jumper pins, but I can try this with some scrap patch wire and let you know. I will leave the JFET on the bottom right corner alone and mounted as is.

Just sending an image to be clear and clarify.
Thanks.

Yes, that is exactly correct. If this test is "positive", then we can assume the JFETs are bad.

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

Test was “positive”.
I desoldered and removed the 3 J111s and connected the “green wires” on A and B with a short wire.
The Blend knob was a dry-wet mix. It went from 100% dry 0% wet all the way to 0% dry 100% wet where I couldn’t hear my played note, only the repeats. Modulation depth and rate still worked as well.

I desoldered those wires and bridged the connections at the red line on C.
This disabled the device. Turning the blend at least to mid and cranking the repeats produced zero effect. No repeats.

So, confirmed. The JFETs were either defective or there was something wrong at the gate pins which forces the toggle in the transistor.
Are the J111s OK with this setup? The stats (eg LDSS) between the J111 and J113 are drastically different and are both listed as valid options in the instructions. Is the voltage at the gate pin (0.01 to -0.05) enough to trigger the switch in the JFET or would a different type be a better option? (Like the J113)
I’m worried about getting replacement J111s and just having the exact same issue.

We've used J111 in these without problems before. Keep in mind, your kit isn't the first and only Echo Royal with J111 in existence. Yours is however, the first time I've seen a problem like this. If I had to guess, all of the JFETs in your kit were exposed to ESD either here during packaging or there while unpacking. If you're worried that the signal from the MCU isn't strong enough, they wouldn't be "open". They'd just stay "closed" because that is their natural resting position. The fact that they are all stuck in the "open" position, even when there's a low signal from the MCU, suggests that they all have an internal short. And the only thing I can think of that would cause that other than exposing them to too much supply voltage is ESD.

Anyhow...I'll be sure to send you something other than J111 just for good measure. I'll test them and package them in anti-stat and hand them over to shipping. They will await your email at sales@buildyourownclone.com

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

Thanks Keith,

I’m wondering if it had something to do with customs at the Canadian border. It was a while ago that I unpackaged it, but I don’t recall the J111s being in a separate anti-stat package. I’m not sure if they scanned or opened the package to inspect the contents.

Let me know if there is anything else you need me to do.

Hey, question…
While I waited for the parts, I removed the other transistor (second A) that I believe you said was for trails.
I then re-jumpered the drain and source like in the test above and tried the same for the other JFET we didn’t touch in attempt to have the effect always on and in trails mode. I didn’t have any wet signal so I removed the jumper on the trails JFET and it’s the same.

If I have dry signal that goes from 100% to 0% with the mix dial but 0% wet always, is that related to my jumper wires maybe not being soldered well or could I have screwed something else up my moving it around and soldering-desoldering things so much??

It's hard to say, but yeah...you probably just didn't jumper something correctly. In "normal" bypass, the A and B JFETs are closed and only the C JFET is open, allowing just dry signal into the mixing output buffer.

In "trails mode" bypass, the two A JFETs are closed and both the B and C JFETs are open. By closing that first A JFET, it blocks any new input signal from going into the delay chips. And it also blocks the dry signal that is affected by the mix knob. But it still allows the output of the delay chips to be mixed back in at the output buffer via the B JFET. So you are still getting echo, but only what was in the feedback loop when you pressed the bypass button, which decays at a rate dependent upon the repeats knob.

So I would assume that didn't jumper that first A JFET that allows signal into the wet signal path.

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

So, that is for bypass modes.
To have it on with trails active I need to have A1 open and A2 open. B1 open for trails and C1 closed?

I removed all of the JFETs so I think my issue is that I didn’t jumper all 4 segments. By default, my faulty JFETs were still letting a signal through one path; now some of mine aren’t connected. That makes sense.

Seems like you’re saying the A JFETs are controlling signal flow to mix, the B is for trails, and C is used in combination to toggle bypass.

No. The first A lets original guitar signal into the delay line. B lets wet signal into the mix.

You need to understand that the difference between what I'll call "mixed dry signal" and "bypass dry signal". C is bypassed dry signal. Mixed dry signal is what goes through the second A JFET. I call it mixed dry signal because it is affected by the mix pot. It can be attenuated. The bypass dry signal is not affected by the mix pot.

So in normal bypass mode, both A JFETs and the B are closed. Only C is allowing bypass dry signal through. In trails mode, C is still allowing bypass dry signal through, but the second A is closed, so no "mixed" dry signal is getting through. B is open, so the repeats are being allowed to get through for as long as the repeats knob is set. But because the first A JFET is close, no new signal is getting into the delay line.

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

Excellent explanation, thank you.
I get it now. There are 2 separate dry signals…. One for bypass to just send the input signal through (C) to the output and one that will allow a copied dry signal into the mix (A1) separate from the wet effect signal (through B) into the mix. The blend knob controlling the ratio of dry (A1) and wet (B) signals that is routed through A2.

I tested with A1 & B open and C and A2 closed. I think my mistake was closing A2 which would prevent the mix from getting through and might explain why I wasn’t hearing any repeats. I was hearing the dry signal (not sure which dry signal - bypass or mixed dry)…. And with having the C and A2 closed to stop the mix, that’s a little confusing…. But this is making more sense now. I’ll troubleshoot again later.
Thank you.

I jumpered A1 and B open to allow the dry and wet signal through to the mix. C is closed to block the dry bypass signal. A2 is open to allow the mix through.
Still not hearing repeats…. Unless I create some crazy noise with my fuzz or some other way. Maybe jumping like this is a bad idea.

Sorry, I’m trying LOL

If you want one of them "closed" simply don't put any jumper at all in.

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

I originally didn’t connect anything but it wasn’t working so I tried to jumper the top and bottom on C to close it. I’ve removed it again.
Great. I think I’ve done something else because I can’t get any repeats at all. And turning the blend all the way up kills all volume.

<sad face>

That means something in the wet signal path (JFET B) is broken. We know it used to work before, so it's probably something associated with the removal of the B JFET or a problem with one of the nearby components. A signal tester would be really useful here. Do you have one of those?

_________________
*patience is a virtue*

Please do not PM me. email is prefered. keith@buildyourownclone.com

I have a multimeter for sure. Let me know where you’d like test results. Until then, I will do my best to test through in ether direction from the B JFET to see if I can find an issue.
It doesn’t look like anything is damaged or disconnected. I’m pretty careful with soldering too not t leave heat on for long and to give time between applying heat again (or to other component legs). Hopefully I didn’t damage some other component.