Hi Forum!
Just got a cheap mini pedal from Mosky called SHO.
It's a super Hard On clone and it's a V1 with the two diodes parallel to the high ohm resistors on the input.
It sounds really nice, but the noise is just terrible. Read somewhere that this is NOT unusual even with the original from Zvex.
Was curious if anyone who build the version without the two diodes, but with a zener between source and gate on the FET could say if they are having noise issues as well? Maybe it could also be a faulty/noisy FET?
I have both FET's and Zeners on order and will experiment when they arrive, but was curious if someone had any experience and could offer some help.
Thanks in advance!
Regards,
Lars
Hi Lars, can you post a link to the modded version, and maybe the original as well (tho original is simple and readily available, I know...)? Pretty sure we get what you mean, but hey, it helps...
I don't really see where "horrible" noise would be built in to this circuit, except that it's amplifying whatever is present on the input. Do you have a gut-shot of the Movsky? Curious if they used shielded cable for the input...building practices can affect noise a bit. SOME noise is to be expected, but not HORRIBLE, if you get my point...some hiss is normal.
Rebuilding the SHO to your taste is an excellent idea! Grab the breadboard...
Mike,
Thanks for the reply. Yes, have included schematics and photos.
I did change the caps to see if maybe it was a faulty one.
Breadboard standing by..
Regards,
Lars
(https://s1.postimg.org/6lcmynigjf/IMG_0146.jpg) (https://postimg.org/image/6lcmynigjf/)
(https://s1.postimg.org/2unhdett1n/IMG_0147.jpg) (https://postimg.org/image/2unhdett1n/)
(https://s1.postimg.org/1kkk73c1gb/IMG_0148.jpg) (https://postimg.org/image/1kkk73c1gb/)
(https://s1.postimg.org/9clp6q5n63/IMG_0149.jpg) (https://postimg.org/image/9clp6q5n63/)
I don't like how grounds are connected in that pedal:
Signal grounds and PCB ground are connected together through jacks-bolts/enclosure/potentiometer-bolt;
Input-jack ground is used for the LED and for grounding the fx-in when the effect is bypassed.
I would re-do it differently, using also the gnd pads already present on the PCB
There isn't much to do with this design a part from reducing the two 10Meg resistors on the MOSFET's Gate to 1Meg and change the input capacitor accordingly (if the input cap was 0.1u then the new one should be 1u in order to have the same corner frequency as before).
You can add a capacitor (~100u) near the power supply, it will reduce the ripples from the power supply even more.
I have built this, version without 1N4148 diodes, and it is not noisy, but it amplify a lot, so any noise your guitar or cable produce will be amplified by this little beast.
Thanks for the answers!
Tried rewiring and it helped some, but there is still too much noise.
When I use my tremolo it sounds like the bottom horn on a leslie amp.
Will try with the zener when they arrive. Also hoping the FET is noisy.
Please let me know if you have any ideas, thanks.
Regards,
Lars
(https://s1.postimg.org/19wza6prmj/IMG_0151.jpg) (https://postimg.org/image/19wza6prmj/)
I think what the others said is good advice. This shouldn't be TOO noisy - is there any other source of noise in your room?
I suppose you are having "Hiss"? Have you powered it from a battery only, using jumper wires? (watch polarity!)
What was suggested about moving the LED ground is good advice - IF it is grounded to the input that may cause problems, so I'd move it to PCB ground point (I take them all back to where batter "-" connects, in a star pattern)....and you could try using shielded cable for input/output connections, with only ONE end of the shield connected to ground; though I am not convinced that coupling is causing your problem. It doesn't hurt to shield, however!
Biggest trouble with a booster is that ANY noise around, at all, is amplified by the active device...that includes any junk coming in on your cable, thermal noise of higher-valued resistors, and any crud that may be in your power supply. I don't use actual "boosters" with effects that should be hiss-free such as trem, for that reason - instead I try to use preamp-type things that are composed of several cumulative stages rather than one stage cranked up, if that makes sense. It keeps things more stable, anyway.
I agree, it's good advise - really nice with all the helpful people around here.
It's really not a hiss, it's lower frequency. There is also his of course and this raises when turning the boost up. But the hiss seems normal and nothing too bad.
Tried with a battery an the noise disappeared - was suprised as I use isolated power supply's for everything in the chain.
Then I tried putting a 220uf cap across the DC jack and this helped some, but not quite enough.
Going to rearrange the led ground today and maybe try a C-R-C filter on the DC input.
Thanks again!
Regards,
Lars
QuoteGoing to rearrange the led ground today and maybe try a C-R-C filter on the DC input.
Proper power supply filtering can make a huge difference, especially in high input impedance circuits. I would go with a 1N5817 diode (for reverse polarity protection) and 100Ω resistor in the +line and a 100-220µF electrolytic with a parallel 10-100nF ceramic or film cap to ground after the resistor. The small cap is necessary because electrolytics tend to have an increasing impedance at high frequencies. That should take care of most of the noise from the power supply.
A slightly off-topic note: I have always found the standard SHO way to hissy. I might be overly sensitive when it comes to noise. One thing is, that n-channel MOSFETs seem to be inherently noisy (mainly high frequency hiss) in high gain applications. Not very noisy, but slightly more than other transistors. So I made a negative ground p-channel version of the SHO using a BS250 and it is
much less hissy. This is not something that can easily be done to an existing circuit (all polarized parts must be turned, some relocated) but if your plan is to build the quietest possible MOSFET booster, I would suggest giving it a try.
Cheers,
Andy
Thanks, Andy - that helped.
Arranged the power so there is 220uf across the DC jack and this feeds the led. Then I took the DC and ran it through a 1N5817 -> 91R and decoupled with 100uf/10nF before feeding the circuit.
The noise is almost gone. Very happy with this.
Thanks everyone!
Regards,
Lars
So, while the noise was all but gone, it felt a little slow under the fingers.
Took out the 100uf/10nf and put in a single 22uf instead and it feels more alive under the fingers.
The noise is still ok.
If you guys hadn't pointed me in the right direction I would be stuck changing FET's and not getting anywhere, so thanks again!
Regards,
Lars
Please excuse me for this OT:
Could someone please explain how to calculate the input impedance of this circuit?
Thanks a lot
Quote from: intripped on October 29, 2017, 10:14:46 AM
Please excuse me for this OT:
Could someone please explain how to calculate the input impedance of this circuit?
Thanks a lot
The input impedance is the gate-to-ground resistor in parallel with the series combination of gate-to-drain resistor and drain resistor.
Zin=Rg||(Rgd+Rd)
Shouldn't the source resistor (pot) also be involved in this calculation?
I've read that the input impedance changes accordingly to how the pot is set
May I ask: is the Mosfet an BS170?
Yes
Quoteis the Mosfet an BS170?
I heard that Zvex use low noise version of these : Zetex if I'm not mistaken.
Over here in Europe, you can get them here:
https://www.banzaimusic.com/BS170P-ZETEX.html (https://www.banzaimusic.com/BS170P-ZETEX.html)
One small nuance - MOSFET transistors themselves are quite noisy.
Perhaps what you are trying to remove is the sound of a transistor that will not go anywhere.
It is enough to look at the rollers of pedals made on these transistors and everything will become clear
1) SHO
2) Box Of Rock
3) Box Of Metal
4) JHS Charlie Brown
5) JHS Prestige
6) Catalinbread RAH
7) EQD Black Eye
8) AMZ Mosfet Booster
Thanks guys. Yeah, so apparently Mosfets are kinda noisy.
But the SHO sounds soo good, it's really worth the little noise it has.
It's not bad anymore, can even run it in front of my gain pedals no problem.
Never thought I would find something better than my old TC Booster/Distortion, but I think I have.
Regards,
Lars
Quote from: alfafalfa on October 30, 2017, 02:13:41 PM
Over here in Europe, you can get them here:
https://www.banzaimusic.com/BS170P-ZETEX.html (https://www.banzaimusic.com/BS170P-ZETEX.html)
Or here: http://www.retroamplis.com/BS170P/en even cheaper ;)
So I just converted to ver2. Can't do A/B for obvious reasons, but it still sounds great. If there is a difference it is small for sure.
There is still a bit of noise and even though it is perfectly acceptable I must admit those low noise Zetex Mosfet's are tempting.
Regards,
Lars
> how to calculate the input impedance of this circuit?
Depends on the gain (due to D-G feedback). Gain depends on resistors and, full-up, MOSFET parameters. Not an easy calc if this is not already clear.
But to simplify: if the gate resistors are 10Meg (a fine value) then the Zin will not drop "too low" for e-guitar. If they are 1Meg, max gain shaves treble.
Thanks PRR for your answer!
I think that treble shaving when gain increases is a desirable effect, I just wanted to predict the amount of this shaving, since it's not always easy to actually hear it.
Also, I would like to better understand why in the box of rock circuit, the boost circuit has got 1M resistors.
(https://effector-guitar.com/wp-content/uploads/2016/02/BOR-1024x773.png)
> why in the box of rock circuit, the boost circuit has got 1M resistors.
They were on sale?