I made this audio probe for testing. I used a 100nf 400v capacitor. It reads on the multimeter 85.5nf
My question is would it be safe to use it or could it damage my amplifier or guitar pickups or something?
(https://i.postimg.cc/WdFkH6zb/IMG-20241019-215612.jpg) (https://postimg.cc/WdFkH6zb)
Not if you use it correctly - question tho - what are you using it on? If you're using it in an actual AMPLIFIER, you need to know the peak voltage of the power supply. I'd use a cap rated at least 20% over that...very high voltages are present in a tube amp, can be over 600V...
For normal 9V/18V use, you should be fine. The VALUE of the cap doesn't matter, it's the voltage rating that's important, as I mentioned.
The lower value the cap, the more trebly the sound will be that you hear, is all...don't use it to determine sound quality ;)
I am sorry i forgot to mention what i am gonna use it for. It's for testing pedals thay run on 9v.
So it's safe to use although it's reading 85.5nf?
Because what i saw online that the lowest you can go is 100nf.
That's why i am asking if it's safe :icon_biggrin:
I'm not sure what Sir Mike is telling but I'm pretty sure he speaks the language of truth.. :icon_wink:
Quote from: Lost_soul on October 19, 2024, 04:25:23 PMwhat i saw online that the lowest you can go is 100nf
It depends on the input impedance of whatever you're plugging into because it forms a high-pass filter with it: https://stompboxelectronics.com/resources/rc-high-pass-filter-response-interactive-calculator/
Again, the sound will not be hi-fi or representative of the actual 'tone' you can expect from the pedal (5-7 points this out), but for testing purposes, yes, you're fine! 15nF difference is probably not very audible even if you Were looking for the actual tone. Things sound different in different stages anyway.
I think I used a 47n/600V cap in mine, in case of use in amplifiers. It's all covered in shrink now so I can't see it to be sure. Works great.
Thank you so much mike. I hope you have an amazing day/night :)
Quote from: Lost_soul on October 19, 2024, 05:45:21 PMThank you so much mike. I hope you have an amazing day/night :)
You too! Welcome to the forum, and I hope you have fun using your audio probe :icon_mrgreen:
I used the audio probe it worked!
But the pedal i was testing it on (zvex SHO) was not working :(
(https://i.postimg.cc/XG1WJY2p/IMG-20241020-151329.jpg) (https://postimg.cc/XG1WJY2p)
In the previous photo, the marked spots didn't produce any sound. What could be causing that?
Btw i used a 2N7000 instead of the Bs170 (i know pinouts are different and i placed it correctly, and i used a b5k potentiometer as reverse log pots aren't available here in egypt.
Thank you.
Quote from: Lost_soul on October 20, 2024, 08:44:10 AMthe marked spots didn't produce any sound
Well one of those points is 9V DC, not in the signal path. There should be
some signal at the source though.
Are you by any chance confusing the power supply input with the signal input? The path you should follow is Input -> 100n input cap (same as gate) -> drain (same as output cap positive) -> output cap negative.
The voltages at the MOSFET's three pins should also reveal if everything's set up properly.
I believe I had an issue like this, with 2N7000 instead of a BS170. I had to increase the drain resistor to get it to bias...perhaps removing the resistor and tacking in a 50k pot would help you dial it in?
First test for signal at the drain. Set gain control at full. If no signal, post voltages at pins.
photos. circuit diagram. voltages. always.
Quote from: Lost_soul on October 19, 2024, 04:25:23 PMBecause what i saw online that the lowest you can go is 100nf.
don't say this out loud, cause we'll be telling you
all kinds of nonsense.
The purpose of audio probes like this is essentially to pinpoint just where, in the signal path, things are not going as intended. To a much lesser extent, they might be used to assess signal quality, but the primary role is to enable one to hear that the signal is present, and that it is, perhaps, louder after a gain stage than before it. As such, the actual capacitance value is of only modest important. As long as it permits enough bandwidth to pass, you'll be okay. If one used, say, a 100pf, rather than 100nf cap, what you hear in your test amp might be so thin and unrepresentative of the signal that you couldn't tell if any given stage was doing something untoward with your signal.
Quote from: duck_arse link
quote author=duck_arse link=msg=1291001 date=1729434626]
photos. circuit diagram. voltages. always.
Quote from: Lost_soul on October 19, 2024, 04:25:23 PMBecause what i saw online that the lowest you can go is 100nf.
don't say this out loud, cause we'll be telling you all kinds of nonsense.
Here are some images.
(https://i.postimg.cc/JHWw6Yz5/IMG20241020182843.jpg) (https://postimg.cc/JHWw6Yz5)
(https://i.postimg.cc/KRxStWVZ/IMG20241020182916.jpg) (https://postimg.cc/KRxStWVZ)
(https://i.postimg.cc/1ns2DnCv/IMG20241020183100.jpg) (https://postimg.cc/1ns2DnCv)
(https://i.postimg.cc/G99zVxyk/IMG-20241020-183231.jpg) (https://postimg.cc/G99zVxyk)
I know the soldering is messy because i was rushing but i made sure there are no bridges.
For the voltage readings:
Drain is +9.07v
Gate is +1.23v
Source is +0.02v
These measures were taken with the gain pot fully clockwise and the toggle switch is on.
Here's the schematic for the circuit:
(https://i.postimg.cc/zHmLdNh2/ZVEX-SHO.png) (https://postimg.cc/zHmLdNh2)
I think this is the same as the stripboard layout.
So what is the problem?
Quote from: Lost_soul on October 20, 2024, 11:49:41 AMSo what is the problem?
The drain voltage should be about 4.5V and yours is 9V, like the MOSFET isn't in there.
(Your gate voltage is likely to be the result of your multimeter loading the 10M+10M voltage divider.)
The problem is the 2N7000 and BS170 have different pinouts. You need to probably need to reverse the device, or at least match-up the device with the layout to confirm a pin-out issue.
Bottom view
2N7000 = \s g d/
BS170 = \d g s/
It seems you know about the pin-out difference.
As it stands the voltages are wrong. What about your pot wiring. Does it really connect to ground?
Can't see from the pic, but is the wiper bridged to the ground end of the pot? The link is broken for me so I can't enlarge it. If the wiper isn't connected to anything, you won't have much of a gain pot, just a straight resistor. Sorry if it is, I'm half blind here.
Quote from: GibsonGM on October 20, 2024, 03:19:43 PMCan't see from the pic, but is the wiper bridged to the ground end of the pot? The link is broken for me so I can't enlarge it. If the wiper isn't connected to anything, you won't have much of a gain pot, just a straight resistor. Sorry if it is, I'm half blind here.
Yes the middle lug is connected to the grounded one as they both are grounded in the layout i shared.
Btw i changed the transistor with another one but i used a socket this time and it worked. It has a lot of noise tho, like white noise or static. :icon_biggrin:
Quote from: Rob Strand on October 20, 2024, 02:41:27 PMQuote from: Lost_soul on October 20, 2024, 11:49:41 AMSo what is the problem?
The drain voltage should be about 4.5V and yours is 9V, like the MOSFET isn't in there.
(Your gate voltage is likely to be the result of your multimeter loading the 10M+10M voltage divider.)
The problem is the 2N7000 and BS170 have different pinouts. You need to probably need to reverse the device, or at least match-up the device with the layout to confirm a pin-out issue.
Bottom view
2N7000 = \s g d/
BS170 = \d g s/
It seems you know about the pin-out difference.
As it stands the voltages are wrong. What about your pot wiring. Does it really connect to ground?
Yes, pot lugs 1 and 2 are connected to ground. I changed the transistor with another one but socketed it (i had to go through 2 more until i finally found a working one, ig parts here in egypt are garbage)
It worked yes But there is high noise floor idk why.
Quote from: Lost_soul on October 20, 2024, 04:15:51 PMYes, pot lugs 1 and 2 are connected to ground. I changed the transistor with another one but socketed it (i had to go through 2 more until i finally found a working one, ig parts here in egypt are garbage)
It worked yes But there is high noise floor idk why.
It's possible the MOSFETs are damaged. That could explain the noise.
I know some people had trouble with 10M resistors so they used 1M. It's possible this gets around
damaged and/or leaky MOSFETS. The circuit behave little differently with 1M.
Another possibility is people have had problems with MOSFET circuits oscillating. Sometimes oscillation can sound like noise. More often there's a weird fizziness to the sound. Here's a few fixes which have worked:
[Image server currently out of service - will post image later]
MOSFET Oscillation Fixes - V10.png
The simplest fix is to add a 100pF from the gate to ground, then perhaps a 1k in series with the input (such that the 100pF is on the gate side).
I've seen another schematic that has the zener connected between the gate and the source. Which is the correct one?
Quote from: MaxPower on October 20, 2024, 09:24:48 PMI've seen another schematic that has the zener connected between the gate and the source. Which is the correct one?
As long as Zener voltage is lower than MosFet V
GS maximun rating plus Source voltage minimum one (i.e. grounded Source) it will make no difference..
To put it another way, for a grounded Source (worst case scenario for max V
GS) you can consider Zener's Anode connected to Source.. :icon_wink:
Note that many zeners have a high junction capacitance that varies with applied voltage. In fact, there was one design of an audio filter I saw that used zeners for tuning by changing the voltage across them. The voltage never got up to the zener regulating voltage, it was just used to change the capacitance, on the order of 1nF to 10nF.
Quote from: Rob Strand on October 20, 2024, 05:16:51 PMIt's possible the MOSFETs are damaged. That could explain the noise.
I know some people had trouble with 10M resistors so they used 1M. It's possible this gets around
damaged and/or leaky MOSFETS. The circuit behave little differently with 1M.
Another possibility is people have had problems with MOSFET circuits oscillating. Sometimes oscillation can sound like noise. More often there's a weird fizziness to the sound. Here's a few fixes which have worked:
[Image server currently out of service - will post image later]
The simplest fix is to add a 100pF from the gate to ground, then perhaps a 1k in series with the input (such that the 100pF is on the gate side).
Yeah probably i got a shitty patch of Mosfets.
For your solution how can i do that and there is no space on the board?
Could the noise be because i got a shitty socket installed and i had to wiggle the transistor until it worked (ie: it's inconsistent)
Maybe i will try to remove the socket and solder the Mosfet directly again but with careful not to burn it.
You could add the cap to the back of the board (solder side) by just stretching it across the rows, from gate to ground. It is helpful to use shrink tubing on the cap legs where they cross other rows to insulate them, of course so that they don't short anything out!
I would prefer to keep the socket, since you may need to use a new mosfet. Replacing the solder connection too many times can be messy and damaging. I have had noisy mosfets happen too, it's very annoying!
Quote from: Lost_soul on October 21, 2024, 07:24:45 AMYeah probably i got a shitty patch of Mosfets.
For your solution how can i do that and there is no space on the board?
Could the noise be because i got a shitty socket installed and i had to wiggle the transistor until it worked (ie: it's inconsistent)
Maybe i will try to remove the socket and solder the Mosfet directly again but with careful not to burn it.
I've added the image to the previous post,
https://www.diystompboxes.com/smfforum/index.php?topic=132552.msg1291019#msg1291019
Yes, the noise could come from many place. Even poorly manufactured FETs (you can even get moisture making its way through the plastic casing can cause issues.)
So while this technique has worked on this circuit there's no guarantee it will fix your problem.
I'd go with with GibsonGM's recommendation. For issues like this where you aren't sure a solution will work for you, it's best to have a "tack-on" solution as an experiment. If you had an oscilloscope you might be able to see the oscillation but without that you have to do the experiment.
can we have the voltages on the working build please? it might just be the mosfet isn't biased as it would like.