AionFX Elysium Troubleshooting/Debugging.

[valveandsound]

Hello everyone, I have been building pedals of all stripes for several years but this one has me really stumped.

Project is the Elysium from AionFX. This is an Ibanez Echomachine EM5 clone.


I assembled the Elysium PCB kit with the supplied components, performed the delay range calibration and confirmed wiring was correct. However I get very weak repeats with no sustain.

Steps I confirmed in building the unit:
1. Verified all resistors were correct value for given location
2. Verified the unit does pass audio cleanly in the dry path and at unity gain from input signal
3. Reflowed all solder joints, checking for solder bridges
4. Verified voltages should be at locations as noted in schematic.

Next thing I did was run a 1Khz tone at 128mvAC into the input.

On Pin 22 of the M65831AP I get around 220-230mvAC signal.

On Pin 15 I get approximately 20mV of signal from what I believe is the chip output.
At the output of the following opamp stage pin 1 i get 38mVAC.

Steps I confirmed after this:

5. Double checked and confirmed all resistor values in this region of the circuit.
6. Ordered another M65831AP chip from AionFX
7. Swapped around the LM833Ns from IC1 to IC2
8. Confirmed oscillator works from roughly 600khz to 4Mhz, which is in spec as per the documentation.
9. Confirmed voltages were correct on M65831AP (confirmed VC on all the required pins listed on the schematic)
10. Lifted R21 from circuit to ensure that nothing downstream was loading the output down.

After the above steps I still get no improvement in the signal. I am at my absolute wits end with this thing.

So, a small request to anyone who might have a completed version of this proejct? if I can make of you? Could you please confirm what kind of output signal I should expect from the M65831AP chip itself for a given input voltage, and what level should I expect from the output of IC2A? I suppose it's possible that I received 2 bad M65831AP chips but this seems incredibly unlikely to me.

Can confirm the dry signal works properly, and the delay range, even when heard faintly, correlates with proper delay times. So the only thing I can conclude is something is amiss with the output of the M65831AP chip.

Any thoughts or comments would be of great help.

I have included an annotated schematic to show where voltages were shown, it is attached.

Below is a link to the documentation from AionFX:

[valveandsound]

https://aionfx.com/app/files/docs/elysium_documentation.pdf

[valveandsound]

[aion]

I can confirm we've never had any individual reports of the delay chip failing, so I would focus elsewhere, mainly the oscillator (everything to the right of the chip in the schematic) which is directly responsible for controlling the chip. If something is wrong with the oscillator, then it will seem like the chip is bad because it's not performing correctly, but the oscillator is the puppetmaster so to speak.

You could also look more at the feedback path, but a signal boost is to be expected from the IC2A stage so I don't think it's that.

[valveandsound]

Thanks for the follow up. These are scope shots from the Xtal pin on the delay chip. Seems like these should look cleaner. Does the oscillator produce a pretty clean sine wave?

[valveandsound]

Thanks for the follow up. These are scope shots from the Xtal pin on the delay chip. Seems like these should look cleaner. Does the oscillator produce a pretty clean sine wave?

[valveandsound]

I went ahead and replaced everything in the oscillator circuit except for resistors, which measured correct. This is what the clock signal looks like now, and at Pin 2 of the IC. Should it have this much ring on the edge?

[valveandsound]

scope shots are here:

https://imgur.com/a/xM8779d

https://imgur.com/a/dlQIxln

[amptramp]

scope shots are here:

https://imgur.com/a/xM8779d

https://imgur.com/a/dlQIxln


It is difficult to avoid overshoots and ringing on a square wave.  If you are fixated on getting rid of it, you could use a pair of diodes biased to conduct when the wave exceeds the normal high and low levels but I doubt the ringing will actually do anything detrimental.

[ElectricDruid]

scope shots are here:

https://imgur.com/a/xM8779d

https://imgur.com/a/dlQIxln

It is difficult to avoid overshoots and ringing on a square wave.  If you are fixated on getting rid of it, you could use a pair of diodes biased to conduct when the wave exceeds the normal high and low levels but I doubt the ringing will actually do anything detrimental.

+1 agree.For a roughly-5MHz waveform, that looks alright to me.

[valveandsound]

Thanks everyone for chiming in. If incoming clock seems clean enough I guess I will move on to looking somewhere else!

[tootsMcgee]

All right, I've got my Elysium open, 128mV @ 1kHz going into input. All measurements taken with knobs at 12

Pin 22: ~200mV (matches yours)
Pin 15: ~220mV (does not match yours)

It could be that you got two bum chips in a row but I agree that it's pretty unlikely. What do your chips look like?

One thing that comes to mind: are the diodes OK? Part of the sound of this pedal are the diodes. If they're shorted then you won't get much gain. Also if C18 is a significantly wrong value it may sound muffled. Just spitballing ideas.

Here's the full complement of measurements:

Delay chip (Multimeter, DC)
Pin 1: 5.15V
Pin 2: 4.42V
Pin 3: 0.49V
Pin 4: 5.15V
Pin 5: 5.15V
Pin 6: 5.15V
Pin 7: 5.15V
Pin 8: 0.13V
Pin 9: 5.15V
Pin 10: 0.13V
Pin 11: 0.13V
Pin 12: 0V

Pin 13: 2.57V
Pin 14: 2.57V
Pin 15: 2.57V
Pin 16: 2.57V
Pin 17: 1.26V
Pin 18: 1.26V
Pin 19: 2.57V
Pin 20: 2.57V
Pin 21: 2.57V
Pin 22: 2.57V
Pin 23: 2.57V
Pin 24: 5.15V

Delay chip, scoped, only pins with AC present, voltages are p-p
Pin 2: 1.56MHz @ 5V (delay knob at 12 o'clock)
Pin 15: 1kHz @ 220mV
Pin 17: Very weird saw wave, my scope won't lock onto it. Appears 2kHz 80mV
Pin 18: Same as above

Pin 21: 1Khz @ 220mV
Pin 22: 1Khz @ 220mV

I don't have any signal at pin 23 but I do have 1kHz signal at lug 2 of the Feedback pot, so I think the filter network in front of the delay chip is to blame.

IC1 DC:
Pin 1: 4.4V
Pin 2: 4.4V
Pin 3: 4.3V
Pin 4: 0V
Pin 5: 4.4V
Pin 6: 4.4V
Pin 7: 4.3V
Pin 8: 8.8V

IC1 AC:
Pin 1: 1kHz @ 128mV
Pin 2: same
Pin 3: same
Pin 7: 1kHz @ 230mV

IC2 DC:
Pin 1: 4.4V
Pin 2: 4.4V
Pin 3: 4.4V
Pin 4: 0V
Pin 5: 4.4V
Pin 6: 4.4V
Pin 7: 4.5V
Pin 8: 8.8V

IC2 AC:
Pin 1: 1kHz @ 288mV
Pin 7: 1kHz @ 100mV

I will say that my clock signal looks like your 2nd scope image (square wave with very thin valleys).

Hopefully this helps.

[valveandsound]

Took DC measurements of the chip last night:

1 - 5V
2 - 4.8V
3 - 0.16V
4 - 5V
5 - 5V
6 - 5V
7 - 5V
8 - .001V
9 - 5V
10 - .001V
11 - .001V
12 - .001V
13 - 2.541V
14 - 2.541V
15 - 2.541V
16 - 2.541V
17 - 0.7V
18 - 0.68V
19 - 2.54V
20 - 2.54V
21 - 2.54V
22 - 2.54V
23 - 2.54V
24 - 5V

[valveandsound]

All right, I've got my Elysium open, 128mV @ 1kHz going into input. All measurements taken with knobs at 12

Pin 22: ~200mV (matches yours)
Pin 15: ~220mV (does not match yours)

It could be that you got two bum chips in a row but I agree that it's pretty unlikely. What do your chips look like?

One thing that comes to mind: are the diodes OK? Part of the sound of this pedal are the diodes. If they're shorted then you won't get much gain. Also if C18 is a significantly wrong value it may sound muffled. Just spitballing ideas.

Here's the full complement of measurements:

Delay chip (Multimeter, DC)
Pin 1: 5.15V
Pin 2: 4.42V
Pin 3: 0.49V
Pin 4: 5.15V
Pin 5: 5.15V
Pin 6: 5.15V
Pin 7: 5.15V
Pin 8: 0.13V
Pin 9: 5.15V
Pin 10: 0.13V
Pin 11: 0.13V
Pin 12: 0V

Pin 13: 2.57V
Pin 14: 2.57V
Pin 15: 2.57V
Pin 16: 2.57V
Pin 17: 1.26V
Pin 18: 1.26V
Pin 19: 2.57V
Pin 20: 2.57V
Pin 21: 2.57V
Pin 22: 2.57V
Pin 23: 2.57V
Pin 24: 5.15V

Delay chip, scoped, only pins with AC present, voltages are p-p
Pin 2: 1.56MHz @ 5V (delay knob at 12 o'clock)
Pin 15: 1kHz @ 220mV
Pin 17: Very weird saw wave, my scope won't lock onto it. Appears 2kHz 80mV
Pin 18: Same as above

Pin 21: 1Khz @ 220mV
Pin 22: 1Khz @ 220mV

I don't have any signal at pin 23 but I do have 1kHz signal at lug 2 of the Feedback pot, so I think the filter network in front of the delay chip is to blame.

IC1 DC:
Pin 1: 4.4V
Pin 2: 4.4V
Pin 3: 4.3V
Pin 4: 0V
Pin 5: 4.4V
Pin 6: 4.4V
Pin 7: 4.3V
Pin 8: 8.8V

IC1 AC:
Pin 1: 1kHz @ 128mV
Pin 2: same
Pin 3: same
Pin 7: 1kHz @ 230mV

IC2 DC:
Pin 1: 4.4V
Pin 2: 4.4V
Pin 3: 4.4V
Pin 4: 0V
Pin 5: 4.4V
Pin 6: 4.4V
Pin 7: 4.5V
Pin 8: 8.8V

IC2 AC:
Pin 1: 1kHz @ 288mV
Pin 7: 1kHz @ 100mV

I will say that my clock signal looks like your 2nd scope image (square wave with very thin valleys).

Hopefully this helps.

Thank you for all this data! This is a big help.

To answer your question about diodes - I removed them from the circuit to see if the output would increase, and it made no real change. So i put new one in place and confirmed they were oriented properly.

Comparing my voltage chart with yours, it seems like the only real discrepancy is in the CC1/CC2 (current control) pins and the pins connected to digital ground (Pins 8, 10, 11). Perhaps the Digital ground has some resistance between it and circuit ground - will investigate.

[tootsMcgee]

I was a little surprised to see my digital ground was slightly high on mine, but I don't know enough about circuit design to know why it would be. I also have no idea what the CC pins do. Maybe Aion can chime in there.

Your chip looks like the one at https://aionfx.com/component/m65831ap/ right? Mine is exact: Mitsubishi logo, model number, some number above it, two inset circles, notch, very clean. I bought mine from Aion as well.

Unfortunately I only bought one so I'm a bit reluctant to do more chaotic things. I was wondering if pulling the chip and *very carefully* bending pin 15 flat horizontal then reinstalling the chip to measure pin 15 without it being connected to the PCB would be useful--don't try it though unless someone chimes in, it was just a thought.

[valveandsound]

I was a little surprised to see my digital ground was slightly high on mine, but I don't know enough about circuit design to know why it would be. I also have no idea what the CC pins do. Maybe Aion can chime in there.

Your chip looks like the one at https://aionfx.com/component/m65831ap/ right? Mine is exact: Mitsubishi logo, model number, some number above it, two inset circles, notch, very clean. I bought mine from Aion as well.

Unfortunately I only bought one so I'm a bit reluctant to do more chaotic things. I was wondering if pulling the chip and *very carefully* bending pin 15 flat horizontal then reinstalling the chip to measure pin 15 without it being connected to the PCB would be useful--don't try it though unless someone chimes in, it was just a thought.

Yes I bought two of these chips from Aion, thinking maybe I did get a bum one. I'm going to replace the pair of 100nF caps tied to the CC pins and see what happens next.

[valveandsound]

I think Pin 15 needs that 30R source resistor in order to pass audio. I removed the 30R from the board and measured at the pin 15 out and it was a bunch of digital looking nonsense. Installed a 30R on the back side of the board and i get a (small) reading out of it.

I think c15 is part of an internal opamp NFB loop on the chip. My gut tells me the low output is from the chip end and not the opamp with the clipping diodes after it.

[valveandsound]

Ok, I think C15 is actually configuring an internal opamp in the Digital chip to act like an integrator EG ramp generator. I *think* what this is doing,  is probably similar to a class D amp whereas the output signal is some sort of square wave and when you run that through an opamp thats configured with a large cap in the -FB loop in lieu of a resistor it "shapes" the waveform output from the chip. This would explain why, when removing the 30R resistor, my scope readings were a bunch of digital looking nonsense. Might try replacing C15 as well and see what happens.

[tootsMcgee]

https://www.diystompboxes.com/smfforum/index.php?topic=86297.msg724260#msg724260

I was trying to figure out exactly what the mod/demod and current control lines do (bear in mind I am not an electronics expert) and this post implies that the capacitor sizes can affect the gain, so it would be worth double checking every single capacitor!