Hi,
I'm new to the forum and to pedal building. I recently got the Aion Fx Cerulean Amp Overdrive kit and made it through the build, however something doesn't seem to be working quite right. The pedal powers on, and produces sounds, however the pedal sounds really gritty or fizzy even with the drive really low. I have another blues breaker style pedal, and my Cerulean build doesn't sound anywhere near as smooth as that one, or what I see in the video on Aion FX's site https://aionfx.com/project/cerulean-amp-overdrive/ (https://aionfx.com/project/cerulean-amp-overdrive/)
I went over the troubleshooting section in the build document, which includes the schematic https://aionfx.com/app/files/docs/cerulean_kit_documentation.pdf (https://aionfx.com/app/files/docs/cerulean_kit_documentation.pdf) and tested the voltages for Q1 and IC1 with the pedal plugged into my 1 Spot CS12.
Q1
PIN | Aion FX Documentation Voltage | My Voltage |
1 | 7.79V | 3.52V |
2 | 1.00V | 3.01V |
3 | 0V | 0V |
IC1
PIN | Aion FX Documentation Voltage | My Voltage |
1 | 4.84V | 4.61V |
2 | 4.84V | 4.78V |
3 | 4.82V | 4.48V |
4 | 0V | 0V |
5 | 4.38V | 2.26V |
6 | 4.83V | 4.58V |
7 | 4.83V | 4.58V |
8 | 9.65V | 9.14V |
After seeing that I have fairly different voltages on pin 1 & 2 for Q1 and Pin 5 for IC1, I took the pedal apart and visually checked to make sure that I didn't accidentally bridge any pins or contacts when soldering. Next I double checked the polarity/directionality of the diodes, LED's, capacitors, IC and transistor and they seemed ok, as best as I can tell. I also when over the documentation again an checked that I had put the correct components all at the right spot and as best as I can tell I did. I also went over all the solder joints again just to make sure and added extra solder if I had any doubt if enough had been applied the first time. I tested the pedal again and got the same results.
Here are some pictures of the build. I'm hoping that someone might be able to help me figure out what's going on so that I can fix it and enjoy using this Pedal.
(https://i.postimg.cc/F7w84wsg/IMG-1819.jpg) (https://postimg.cc/F7w84wsg)
(https://i.postimg.cc/1gQdg6F8/IMG-1820.jpg) (https://postimg.cc/1gQdg6F8)
(https://i.postimg.cc/LhDmm9rk/IMG-1821.jpg) (https://postimg.cc/LhDmm9rk)
(https://i.postimg.cc/68kXMyvK/IMG-1822.jpg) (https://postimg.cc/68kXMyvK)
(https://i.postimg.cc/5jsWZPCn/IMG-1823.jpg) (https://postimg.cc/5jsWZPCn)
(https://i.postimg.cc/QFTLKhny/IMG-1824.jpg) (https://postimg.cc/QFTLKhny)
(https://i.postimg.cc/MvXJxy0x/IMG-1825.jpg) (https://postimg.cc/MvXJxy0x)
(https://i.postimg.cc/jnS0DfGH/IMG-1826.jpg) (https://postimg.cc/jnS0DfGH)
(https://i.postimg.cc/8JDVV94n/IMG-1827.jpg) (https://postimg.cc/8JDVV94n)
(https://i.postimg.cc/vDYd2Rc4/IMG-1828.jpg) (https://postimg.cc/vDYd2Rc4)
(https://i.postimg.cc/CBNpM76M/IMG-1829.jpg) (https://postimg.cc/CBNpM76M)
(https://i.postimg.cc/r0yYjkQc/IMG-1830.jpg) (https://postimg.cc/r0yYjkQc)
(https://i.postimg.cc/G925sb4x/IMG-1831.jpg) (https://postimg.cc/G925sb4x)
(https://i.postimg.cc/QBJyctg0/IMG-1832.jpg) (https://postimg.cc/QBJyctg0)
(https://i.postimg.cc/xcXhB9Y1/IMG-1833.jpg) (https://postimg.cc/xcXhB9Y1)
Hi & Welcome.. :icon_wink:
Your Vref seems OK (IC pin 3) so for the lower enough measurement on pin 5 the suspect is your meter's low (slightly over 1M, judging from 2.26V instead of 4.48V) impedance..
(it's effectively set in parallel with R1 resulting into significantly lowering equivalent resistance..)
Q1 measurements make no sense.. :icon_wink:
Check both RX4 (22k) and RX5 (12k) values..
Sorry if this is a noob questions, but just to make sure, when testing the values for RX4 (22k) and RX5 (12k) I would need to desolder them and test them outside of the circuit right?
Also when you say "so for the lower enough measurement on pin 5 the suspect is your meter's low (slightly over 1M, judging from 2.26V instead of 4.48V) impedance.." what do you mean by my meter's low? like do you mean that my multimeter might be the issue? I'm not 100% sure on the model, but it's a mastercraft one, I picked up at Canadian Tire years ago.
(https://i.postimg.cc/MnRtRgsp/64c05cf7a3f821e35bc0c1cc52e1a870.jpg) (https://postimg.cc/MnRtRgsp)
Just lift one leg and measure at 20k range for RX5 and at 200k range for RX4..
All DC voltage measurements (for particular circuit) should be taken with DMM set on 20V range..
Thanks. I tested those those resistors and I'm getting 22k for RX4 and 11.99K for RX5.
I think I found a typo in the documentation however. Page 4 lists 47pF (marked "470") under the MLCC Caps, but page 14 says C2 47pF MLCC (may also be marked "471")....One says 470 and the other 471...I double checked the one I got with the kit and it says 470, which from my understanding is 47pF
The schematic also lists 47pF for C2.
BTW the part number for Q1 that I got with the kit is PN4303, in case that helps at all, since two possible part numbers where listed.
Any other suggestions for things to test?
The thing about "your meter's low impedance:
If your meter had an infinitely high impedance it would not affect the voltage you are measuring. It in real life acts like a big resistor, maybe 1M, in parallel with what ever is there. When you measure the voltage at the input of the opamp, you actually drag it down a bit. If you could measure the other pins at the same time you are doing this, you would see them follow. The other two pins of the opamp are however "buffered" in that when your meter tries to drag them down, the opamp keeps them up.
So: very common thing. you check voltages on opamp, and the input getting the bias reads low, but the other input and the output all read 4.5 more or less.
I took some other voltage readings in case it helps.
IC Pin 1 4.58V
R7 4.58V
R8 4.58V (with tone knob set all the way to the right)
RX2 0V
Measuring with the pcb's mounted in the enclosure and the input/output/DC jacks facing up/top
RX4 9.17V on the left of the resistor and 3.54V on the right side of the resistor
RX5 0V top side of the resistor and 3.01V on the bottom side of the resistor
RX6 2.98V on the left of the resistor and 3.01V on the right side of the resistor
@idy thanks for the explanation. Just so I know, if I where to get a better multimeter with a bigger resistor, would this difference be lessened in order to get better readings? Just wondering since the Aion FX documentation shows 4.38V for IC1 Pin 5, so i'm just curious how they got that result? Or would simply opening up my multimeter and putting a larger resistor in place of the 1M ohm resistor be another option?
I mean the meter has a quality called impedance, which is "like a resistor."
You can't change your meter to improve this, It is likely good enough.
I was just saying, it is normal for the + input of an opamp to read a bit low. If the other two pins (- input and output) are correct, we know to take that other reading in context. We know that, when we aren't measuring it, the + input is the same as the others. (Also when we are measuring it, and "loading" it or drawing it lower, the other two pins follow it down.)
(https://i.postimg.cc/nC1fBKnP/Screen-Shot-2022-05-12-at-11-14-17-PM.png) (https://postimg.cc/nC1fBKnP)
QuoteI'm getting 22k for RX4 and 11.99K for RX5.
You know those should be 10k and 220k? Are the color codes right? I can't see how those numbers could be. You are turning power off before testing resistors? You should. Any reason for the x's? RX4? You mean R4, right?
No, you can't be measuring 0v on one side of R5, unless it is not attached. It goes between - input and output of opamp, pins 1 and 2, both are mid voltage. how can it read 0? did you remove it from the board? Is it soldered in?
Quote from: idy on May 13, 2022, 02:27:25 AM
You know those should be 10k and 220k?
Drain resistor (RX4) should be 22k and Source resistor (RX) should be 12k, so they measure right.. :icon_wink:
Also, 0V at Q1 Gate (at any leg of RX2) is fine..
Still can't get the reason for Q1 particular measurements..
@jrbilodeau: Try to jumper (via an aligator clip) Tone pot lug 3 to OUT (Volume pot wiper) while disconnecting HARD switch middle lug from VR..
(you should have a loud distorted signal..)
Then connect again HARD switch middle lug to VR..
(you should have a more distorted signal of lower level..)
(https://i.imgur.com/OqfdKvB.png)
Sorry, now I see the RXs. "Never mind."
Now that I am looking at the right end of the horse..
It looks like the schematic shows two options, but I see in the docs they don't explain this to you, just say leave CX5 off.
The PCB allow two options:
either you populate C8 and CX5 to get a "stockfish" blues breaker
or you populate C2 and Q1 and you get a morning glory-is thing.
Trouble if you do both maybe? a big feedback loop...
About the 47p cap: 47p is the correct value, and that would be written either "47" or confusingly "470". Yes, "471" means 470, and it looks like the docs have a typo where they list the part both ways...
Q1 seems the big bug to me. Either you have the JFET gate where the drain should be or it's a fake (not unknown) and probably actually a PNP transistor or it's just faulty.
I guess it's also possible that CX4 is shorted letting DC from the preceding stage into the JFET gate.
Quote from: jrbilodeau on May 12, 2022, 11:45:36 PM
I think I found a typo in the documentation however. Page 4 lists 47pF (marked "470") under the MLCC Caps, but page 14 says C2 47pF MLCC (may also be marked "471")....One says 470 and the other 471...I double checked the one I got with the kit and it says 470, which from my understanding is 47pF
The schematic also lists 47pF for C2.
I'm not entirely sure this is 100% correct. find the spec sheet for the manufacturor of your part, see what standards are in use for the markings. 47pF should be marked as 47 or 47pF [or 47R0 in some cases] but I think 470 is illegal. 47K would include the tolerance. 470pF should be marked
either 470p or 470pF, or 471. 0r n47.
if you get the thing working, but sounding "dark" or low gain, you can test-lift this cap, see what it effects.
Quote from: idy on May 13, 2022, 10:26:24 AM
The PCB allow two options:
either you populate C8 and CX5 to get a "stockfish" blues breaker
or you populate C2 and Q1 and you get a morning glory-is thing.
Trouble if you do both maybe? a big feedback loop...
About the 47p cap: 47p is the correct value, and that would be written either "47" or confusingly "470". Yes, "471" means 470, and it looks like the docs have a typo where they list the part both ways...
I went with the morning glory-is option and can confirm that nothing is hooked up to C8 or C5. Also thanks for confirming about the typo. I reported it through the Aion Fx contact form, so hopefully they'll fix that in their documentation.
Quote from: duck_arse on May 13, 2022, 11:06:15 AM
I'm not entirely sure this is 100% correct. find the spec sheet for the manufacturor of your part, see what standards are in use for the markings. 47pF should be marked as 47 or 47pF [or 47R0 in some cases] but I think 470 is illegal. 47K would include the tolerance. 470pF should be marked either 470p or 470pF, or 471. 0r n47.
if you get the thing working, but sounding "dark" or low gain, you can test-lift this cap, see what it effects.
I got this as a kit through aion fx's site, and from what i've see with various forums they seem pretty legit and use quality components. When first looking into the possible typo, I looked up how to read the values on capacitor, as is seemed that there were a few standards. From what I read on various site and video's the ones that use 3 numbers, the first two are the capacitor value in PF and the 3rd number is the multiplier or number of additional 0's to add to the value. So 47pF and add 0 0's = 47pF, where as if the value was 471 the capacitor value would be 470pF.... at least that's how I understood it.
Quote from: anotherjim on May 13, 2022, 10:58:29 AM
Q1 seems the big bug to me. Either you have the JFET gate where the drain should be or it's a fake (not unknown) and probably actually a PNP transistor or it's just faulty.
I guess it's also possible that CX4 is shorted letting DC from the preceding stage into the JFET gate.
To check if CX4 is shorted would I do a continuity test between the two legs of the capacitor? If so I set my meter to 200K ohms and got a reading that fluctuated around 100-111K
The Transistor came with the kit ordered directly from Aion FX and has the following numbers on it FB426 PN4303
After work i'll do the test that @antonis suggested and let you guys know the result. If we don't get anywhere I can always order another transistor...or is there some way that I can take it out of the circuit and test if it's good?
Pn4303 is obsolete but is the right kind and should be pinned DSG. If your meter has a diode test, Red on 3 (G) should give a reading with Black on either 1 or 2. Black on 3 should not read with Red on 1 or 2. This ought to work with the transistor in the board.
Don't do any diode or resistance checks with power on the board!
What Jim said but with resistance measurement.. :icon_wink:
https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/ (https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/)
Quote from: antonis on May 13, 2022, 07:43:49 AM
Still can't get the reason for Q1 particular measurements..
@jrbilodeau: Try to jumper (via an aligator clip) Tone pot lug 3 to OUT (Volume pot wiper) while disconnecting HARD switch middle lug from VR..
(you should have a loud distorted signal..)
Then connect again HARD switch middle lug to VR..
(you should have a more distorted signal of lower level..)
(https://i.imgur.com/OqfdKvB.png)
I unsoldered the middle pin for the HARD switch and hooked up the aligator clip from pin 3 on the tone pot to pin 2 on the volume pot, and I had a loud signal with some distortion, but way less then before. Next I resoldered the middle pin on the switch and did the test again with the aligator clips still on, and it was more distorted then the previous test, but less than without the clip. The volume sounded similar, maybe a bit less.
I also tested the JFET with red on the gate and black on pin 1 and then pin 2 and it seemed ok. I tried the reverse and got nothing which i believe is a good thing. I also did a test shoring source and gate with the red lead and black on the drain and got 278 ohms.
It looks like the JFET is ok, so you need to find out why there is other than 0v on its gate pin. The 1M RX3 is the only thing supplying 0v to the gate (yes, 0v ground is a voltage!). Cap CX2 is the only thing blocking any other DC voltage from interfering with the gate voltage.
As RX3 is very high resistance, it doesn't take much for leakage from parts nearby to interfere with it. Even an unsuitable soldering flux can conduct enough to interfere. An idea is to scrub the area with IPA (alcohol, not beer).
Normally, you would trace this kind of fault by selective disconnection of nearby parts and see if the voltage then corrects. If you have a double-sided PCB this can cause damage as it needs more heat to flow the solder and you can disturb the thru ferrule in the hole that is there to connect both sides of the traces.
I tried cleaning the board with 99% isopropyl alcohol, but i'm still getting the same results.
Regarding the 0V on the gate. If the JEFT is DSG, than I thought I was already getting 0V on the Gate, based on my original readings. It was more the voltages on the drain and source that seemed off.
Q1
PIN Aion FX Documentation Voltage My Voltage
1 7.79V 3.52V
2 1.00V 3.01V
3 0V 0V
Not sure if I'm misunderstanding something, as I'm a bit new at this and learning...but I appreciate the help :)
Also just so I know, is it usually a good practice when building a circuit to go over and clean the pcb with rubbing alcohol afterward soldering to clean off the flux?
Sorry, I must have read crosseyed. Thought there was 3v or so on the gate! But then I didn't know what 1,2,3 referred to and made an assumption!
I'm not sure I like the example voltage of near 7.8v on the drain unless it's meant to distort, but JFET performance varies all over the place from part to part and your example might actually be better biased than Aions was.
Hmm, yeah I'm not sure. The pedal definitely isn't sounding right, even at low volume and low drive there's a bunch of distortion and it sound kind of crackly. It definitely isn't sounding much like a blues breaker. Maybe I should unsolder the component and test them individually to make sure all the values are correct?
I've gone over the schematic and tested all the point listed and ground and am getting 0V and all the points that read VR are coming up with the same voltage...same with points listed as VA.
My current multimeter doesn't have a setting to test capacitors, so I'm thinking of picking up a new one. I read that the ANENG 8008 and 8009 are pretty good for the price. As far as I can tell, the only difference seems to be 8008 has a square wave generator where as 8009 has non contact voltage detection. Would either of these features actually prove useful in pedal building? If you guys have any better recommendations I'm up for suggestions.
Thanks
I was just doing a search in the forum and can across this post that sounds really similar to mine. https://www.diystompboxes.com/smfforum/index.php?topic=125642.msg1197707#msg1197707 (https://www.diystompboxes.com/smfforum/index.php?topic=125642.msg1197707#msg1197707). The voltages that he listed are quite close to what I'm getting and I did some testing and when with the hard and soft clipping switches and I also get very little difference regardless of their position. He mentioned having a bad transistor, and I know that I did the test meantioned in the earlier post and it seemed to come back as expected, but is it possible that it's still bad? before going further, would it be wise to order a new transistor and try that to see if it fixes thing?
Buying another transistor from reputable source is not a bad idea. Maybe you are thinking of an order for your next project?
If you are so suspicious of the Q you could carefully remove it and try to breadboard the booster stage to see if it works. All those components between CX2 and CX4. Nice simple project, FET booster.
I don't see that you have tried a signal probe yet.(?) Necessary and simple implement! Always good to try that before you get all medieval with the soldering iron. I would not dream of blaming a component until I have signal traced a circuit. But FETs are dodgy.
Looked up a few guides on building a audio probe, and I found a .1 uF cap in some old electronics stuff I had from back in my college days. I'll try building an audio probe and see what results I get, and then see about getting another JFET if need be. I found a MPF102 N channel JFET in my old parts kit. Although it's not an equivalent replacement for the 2N4303 or 2N5457 which the kit lists as parts for Q1, would swapping this be a possible test until I can get the right part shipped....depending on how the audio probe goes first that is.
Also if I was to get a better multimeter what do you guy's think of the ANENG 8008 or 8009? My current Mastercraft one doesn't have a capacitor reading setting, and the diode test, doesn't have a audible beep, it seems to be more just on the 2K resistor setting.
Late to the party (because that's what cool kids do, right?)
Go with AN8008. I bought one to replace my big MM since it didn't have a frequency setting. Been going strong for years. 4-digits and inexpensive. Can't go wrong.
However no transistor test sockets if that concerns you.
For your jfets, just be mindful of the pinouts. When used as a gate most jfets usually work however if used as a source of variable current you may have to adjust the bias to accommodate. Spectrum behavior may also differ in that case.
Just something to chew on, likely not going to be a problem for you :icon_smile:
So I made an audio probe tonight and did my best to trace it (I'm still new at this). Things sound pretty good until I got to the the drain or source on the JFET, or at RX4 or RX5. Then everything got all crackly. I had a bit of a hard time figuring out how to properly trace the hard clipping diode section, as I got audio on both sides all of the diodes in the hard clipping section that ranged in volume and amount of distortion. This test was done with the switch in the middle position. Also with the switch in that position I got audio reading on all 3 lugs. The top one was the loudest, the middle one was really faint, but could still be heard and the bottom had slightly less volume that the top one.
With the hard clipping switch set to the top position, I got audio at the 1st lug, and very faint audio at lugs 2 & 3
With the hard clipping switch set to the bottom position, I got audio at the 3rd lug, and very faint audio at lugs 1 & 3
The voltage that i'm getting at VA is 9.14V in front of RX4 (22K resistor), but after crossing the resistor it drops to 3.52V
Btw, do you guys have any good ressources that you'd recommend to learn how to understand how these type of audio circuits work? I took a few courses back in college 15+ years ago as part of my computer programm, but went more into the computer side of things, and don't remember alot of this. I have some basic understanding of things like resistors, capacitors, diodes, opamps, but when you put it all together and how the various parts all interact with one another and how they affect the tone, thats a bit over my head. I'd like to be able to better understand how the circuits work and understand how to troubleshoot, and tinker with mods and stuff. I realize that it'll be a learning process, so just curious on good ressources to help get the learning started.
Quote from: jrbilodeau on May 18, 2022, 12:03:34 AM
So I made an audio probe tonight and did my best to trace it (I'm still new at this). Things sound pretty good until I got to the the drain or source on the JFET, or at RX4 or RX5. Then everything got all crackly.
I don't like that Q1 CS amp fancy configuration with all those NFB items..
Personally, I'd delete CX5, C8 & RX1 and make RX2 an order of magnitude lower..
(but my aspect doesn't help with your issue..) :icon_wink:
CX5 and C8 are used in other optional for this build, but since i'm following the kit documentation for the morning glory style circuit i'm not using them. The non kit documentation for the pcb is found here https://aionfx.com/app/files/docs/cerulean_documentation.pdf (https://aionfx.com/app/files/docs/cerulean_documentation.pdf)
QuoteBUILD NOTES
Morning Glory Gain Stage
The JHS Morning Glory adds another gain stage to the end of the circuit for more available output
volume. You can choose to build the pedal with this gain stage or without.
To build with the gain stage, omit C8 and include RX1-6, CX2-4 and Q1.
To build without the gain stage, include C8 and omit RX1-6, CX2-4 and Q1.
King of Tone
The Analogman King of Tone is two Bluesbreakers in one box. All of the modifications from the King of
Tone are available as part of this project (soft clipping, hard clipping, and Presence control).
In the parts list CX5 is listed as being part of the King of Tone variant, but to omit it in the morning glory build...so they show up in the schematic, but aren't used in the build that I built.
Could you plz also omit CX5..??
(Volume pot lug 3 left connected to CX4 negative plate only..)
That's what I'm trying to explain, C8 and CX5 where both omitted from the build that I did. I've never put any components in that location since I was following the morning glory style build, which didn't use them.
(https://i.postimg.cc/21BbrWBc/Untitled.jpg) (https://postimg.cc/21BbrWBc)
QuoteBtw, do you guys have any good ressources that you'd recommend to learn how to understand how these type of audio circuits work?
Start with the links at the top of the page. GEOFEX has some great lessons called
"technology of...." Great walk through tubescreamer and a few others. Sadly no Bluesbreaker....
When I first saw it, I thought they added a JFET stage to boost the level which with only the single set of hard clipping diodes can seem too quiet for a lot of players, but then they have those caps around it (1uF with 100n in parallel makes no sense). As the JFET stage is inverting, it will cancel out some important frequencies and I'd have to hear it to believe it won't sound horrible! Of course, Antonis pointed this out earlier but he sees it as feedback while I see it as bypass. Anyway, definitely take CX5 and C8 out of there!
That whole JFET stage appears to be some version of a guitar "Fetzer" preamp circuit (68k input resistor lends credence) stuck under the original straight output with the output caps left misplaced on the schematic.
I didn't add anything to CX5 or C8, so there's nothing to take out.
OK.. You said that you have "clear" signal till Drain/RX4 and Source/RX5/RX6 joints..
What puzzles me is the later (Source) crackling 'cause at that point you should have the same signal "quality" as with that at Gate (RX2/RX3 joint) - just a bit of lower amplitude..
But with a quiescent VDS voltage of about 500mV (3.5 -3.01) things get complicated a bit..
Could you plz repeat audio probing with clipping switch set at HARD position (onnly a pair of antiparallel diodes activated)..??
(I'm mainly interested about Source signal, to figure out if your issue relies upon inproper bias or on JFET itself..)
P.S.
For your understanding: Q1 configuration may be considered either Common Source amplifier or Common Drain voltage follower, depending on signal output point of view..
For the later configuration, outcoming signal (from JFET Source) should be almost identical to the Gate incoming one (in phase but of slightly lower volume..)
The above only stand for VDS higher than incoming signal amplitude..
(not even for equal to, for reasons beyond present scope.. :icon_wink:)
For any clipping diode configuration other than D8/D9 pair, Q1 input signal amplitude is higher (considering no losses at Tone and Presence configurations) than allowable for undistorted Source outputVDS..
Any chance you could confirm that swapping the transistor solved your problem? I'm having nearly exactly the same problem as you had and can't quite seem to figure out why. I'm starting to suspect that the transistor might be bad based on what you and the other guy have said.
If swapping Q1 fixed it, could you also let me know what you swapped to?