SHO issues - Beginner alert, help gratefully received

Sibob · Post by **Sibob** » 07 Oct 2013, 17:51

I've built this SHO clone: http://pedalparts.co.uk/wp-content/uplo ... /Boner.pdf
And have another exactly the same ready to box up.
The built one has an almighty pop when the switch is engaged, I'm told that this circuit is particularly prone to this, so I've made up a the AMZ LED pop circuit on vero to try in the next one, hopefully this will work and I'll build one into the finished SHO.
I'm also getting a LOT of hiss though, I understand that this is a high-gain circuit, but is there anything I can do to minimise this?
Here's a 'guts' shot for you to critique........gently

Thanks in advance
Si

jymaze · Post by **jymaze** » 07 Oct 2013, 20:03

1) Against the " pop" : Add a resistor from the IN point to ground. Something between 2.2 and 4.7 meg should be fine in order not to load your pickups too much. If desperate, 1 meg will work but you may loose some sparkle.

2) It is a low gain circuit actually (about 20dB), it should be silent. I can see the jack on the top of the picture is not grounded ---> GROUND IT! Also, are all your ground point tied together? I cannot see the ground from that switch board being linked to the main board or to the power input... I assume it is! If not... You know what to do!

Sibob · Post by **Sibob** » 07 Oct 2013, 21:11

Hi there,

Thanks for the reply.
Great will try the resistor from In to ground.
Ground wise, there is a ground wire from the switch-board to mains socket & IN jack, that's tied to the ground wire from mains to PCB.
Output is not grounded by wire as I thought it would be grounded by the enclosure?!

Thanks again
Si

friedtransistor · Post by **friedtransistor** » 07 Oct 2013, 23:19

Yeah, I think those jacks do have a chassis ground, but I would personally ground both jacks just to be safe. Also, is your input jack doing the stereo switching thing? Because it looks a little odd to me. And just a tip: I would not only ground the jack, but also ground the chassis to the board. That'll reduce humming, and give your jacks another ground path in case the wires snap.

Sibob · Post by **Sibob** » 07 Oct 2013, 23:35

No the input jack is just wired mono as I've no battery option on this.
Ok, great will try redoing the grounds to include output & enclosure.
Will let you know how I get on, thanks for your time

Si

galmar · Post by **galmar** » 08 Oct 2013, 17:35

friedtransistor wrote:Yeah, I think those jacks do have a chassis ground, but I would personally ground both jacks just to be safe. Also, is your input jack doing the stereo switching thing? Because it looks a little odd to me. And just a tip: I would not only ground the jack, but also ground the chassis to the board. That'll reduce humming, and give your jacks another ground path in case the wires snap.

These jacks make contact with the chassis. If you connect a board ground to each of them, leaving them connected on the chassis too, you will form a ground loop. Asking for noise trouble. Right now, pedal ground is connected to chassis via the input jack.

Best option is to use isolated jacks, and tie ground at the chassis at the input with a screw (check for Safety Earth Bond in google). That gives you shielding plus no ground loops.

What types of resistors are you using? Normally hiss is due to thermal noise. High-value resistors are more prone to it, and metal film quieter than carbon.

Sibob · Post by **Sibob** » 08 Oct 2013, 19:37

Thanks for your input.
I'll look into isolated jacks.
The resistors on this built SHO are carbon, the resistors on the SHO ready to box up are metal.
I know the cables in the pictured one are quite long, so I'll box up the new one with shorter cables and with the LED popping vero, see how we get on

Cheers
Si

galmar · Post by **galmar** » 09 Oct 2013, 08:32

Yes try changing the resistors to metal film from carbon film, this should reduce your hiss.

deltafred · Post by **deltafred** » 09 Oct 2013, 18:06

galmar wrote: These jacks make contact with the chassis. If you connect a board ground to each of them, leaving them connected on the chassis too, you will form a ground loop. Asking for noise trouble. Right now, pedal ground is connected to chassis via the input jack.

Sorry but this is incorrect.

Ground (hum) loops occur when you have current flowing along a signal earth/ground. The voltage across any conductor will follow ohms law - Volts = Amps * resistance.

When you have several parallel connections serving the same purpose it can only reduce resistance (and impedance) of the connection.

Where hum loops can be a problem is in power amps where high power supply currents get mixed with signal ground connections unless great care is taken to keep them separate.

galmar · Post by **galmar** » 09 Oct 2013, 20:45

No, a loop is formed. I believe you don't call it ground loop because you think it may not cause the typical trouble followed by that word - it certainly won't, but it is bad electronic practice.

Take a noise source that creates magnetic fields. Create a loop. It will induce current into it, and this will induce noise voltages.

From a parallel connection point of view, what you say isn't exactly right, because the first path is the chassis (your first resistor), but the second one is not a straight wire from one jack to the other - the path through the pcb.

Anyway, it is bad practice to connect twice any ground. Have you heard about the ground lift switches? The same mechanism acting there. The first path is earth, the second path is the series connections of audio grounds that are eventually connected to earth.

EDIT: I attach this link as a reference: http://www.valvewizard.co.uk/Grounding.pdf. Check Figure 15.3.a at page 265. I think it bears great resemblance to our situation.

deltafred · Post by **deltafred** » 09 Oct 2013, 21:38

galmar wrote:No, a loop is formed. I believe you don't call it ground loop because you think it may not cause the typical trouble followed by that word - it certainly won't, but it is bad electronic practice.

Take a noise source that creates magnetic fields. Create a loop. It will induce current into it, and this will induce noise voltages.

From a parallel connection point of view, what you say isn't exactly right, because the first path is the chassis (your first resistor), but the second one is not a straight wire from one jack to the other - the path through the pcb.

Anyway, it is bad practice to connect twice any ground. Have you heard about the ground lift switches? The same mechanism acting there. The first path is earth, the second path is the series connections of audio grounds that are eventually connected to earth.

EDIT: I attach this link as a reference: http://www.valvewizard.co.uk/Grounding.pdf. Check Figure 15.3.a at page 265. I think it bears great resemblance to our situation.

I agree a loop is formed and if there is magnetic interference then it will induce a voltage but I have had more trouble with electrolytic corrosion due to dissimilar metals (aluminium chassis and whatever plating is on the jacks) than I have ever had with induced interference.

Yes it is similar to fig. 15.3 except that is a big assed valve with unknown heater current, probably AC, flowing every which way in the chassis. Valve amps as you no doubt know need very careful grounding. Stomp boxes rarely have more than a few milliamps DC flowing anywhere.

Ground lift switches were outlawed years ago in the UK for very good reason.

galmar · Post by **galmar** » 09 Oct 2013, 22:41

Correct! Of course it may have no audible consequence - still it is just bad practice, that is why I highlight it!

Of course, when saying ground lift switch, I refer to a switch that disengages audio ground from chassis - the chassis should always remain at earth potential! The link also includes this situation somewhere. When applied well, audio ground plus every exposed metal shall be at earth potential, without creating loops. It is somehow the case with pedals - your audio ground, your strings' potential that is, is connected to earth at the amp chassis via your cables. So, there are two ways of properly adding earth potential to your pedal's aluminum case: (1) connect it directly to earth via a mains cable or (2) connect it to the pedal's audio ground, that is at earth potential as we said. So, typical pedals encorporate an open-frame jack that automatically brings the case's potential to earth. (But if you suddenly attach the enclosure to earth via a mains connection, a ground loop is formed! Never seen that though.)

That means that for safety plus shielding considerations, when using isolated jacks, audio ground should be connected to the metal enclosure, normally at the input.

Sibob · Post by **Sibob** » 10 Oct 2013, 09:49

Have I split opinion here?

I've boxed up the metal-film version of the SHO, which also has the AMZ LED pop circuit on some vero.
Carbon-film resistor PCB = left, metal-film = right:

I finished late last night, and so whilst it works, I'll have to wait until this evening to test the gig volume differences in pops and noise floors.
I did notice that Zvex do give notice of the pops on the this circuit, they do seem to dicipate after mashing the footswitch a few times, as long as you don't adjust the level......set and leave

.

I don't think the hiss/noise I'm getting is a grounding issue, it just seems to be the noise-floor of the effect?!

Thanks for all the help so far, will report back later

Cheers
Si

Nocentelli · Post by **Nocentelli** » 10 Oct 2013, 10:14

Sibob wrote:Have I split opinion here?

I don't think the hiss/noise I'm getting is a grounding issue, it just seems to be the noise-floor of the effect?!

Those are my thoughts too. Ground loops are often characterised as causing a "hum" rather than "hiss": Hiss (to me) implies high frequency noise already in the signal chain from the cables/pick-ups etc being amplified by the boosting effect of the SHO circuit. It isn't always apparent in some other boost designs (e.g. EHX LPB-1), maybe due to the lower input impedance loading the guitar and causing some treble loss which would mask the hiss somewhat. The SHO has a very high input impedance (as previously noted), giving a glassy/clear/hifi boosted sound, and might just be amplifying the full frequency range that enters the input with very little roll-off.

galmar · Post by **galmar** » 10 Oct 2013, 10:45

Yes, the conversation above was not supposed to imply that you problem is grounding. I just noted all these because your grounding seems odd - but nobody said it will certainly cause hiss. If I see well though, you attach only the hot wire to the output jack, which means that you ground it via the chassis. I consider this to be bad practice and maybe it picks up high frequency noise too - the enclosure provides a low impedance path for that, and since it can pick it up, it can intoduce it into your audio ground too. The whole chassis could be regarded as a closed loop, a portion of which is the input to output ground connection. So if noise currents circulate within the case, they can do it across your audio ground.

I have never read or investigated this possibility, this is just my assumption. So I again suggest one the following, in this order:
(1) Use isolated jacks and connect audio ground to each of them. Tie audio ground to the chassis at your input too.
(2) You can also use an isolated jack only for the output jack. Then you can connect audio ground to this jack from the pcb. Only drawback: You lose some aesthetics, since from the outside your jacks will look different. I don't know if you mind.
(3) Try to isolate your output jack (or both, but if so attach audio ground to the chassis) by means of using plastic (insulating) nuts. This means that the chassis jack hole must be big enough so that the jack doesn't make contact to your case.
(4) Connect audio ground to the output jack, even if you use these jacks. Notice if this makes any difference. Try this option leaving the output jack unconnected to your chassis too - this is actually option (3), but if you don't have a means of isolating this jack, just leave it "on air" to test whether noise is reduced, and if it does, follow one of the above solutions.

Hiss may indeed be just the noise floor of your pedal plus possible cable noise, as Nocentelli stated. Try the above and see how it responds. And, of course, compare the carbon vs metal film version and tell us if you see any difference.

friedtransistor · Post by **friedtransistor** » 10 Oct 2013, 20:11

Hey, I just thought of something. A few days ago, I made an amp simulator, that I modded. The eighteen from runoffgroove. Well, my version was super noisy. I noticed that if I put a 47n cap on the output fet, it almost eliminated the noise. For you, just add a 47n cap from the drain of the bs170 to ground. It cuts off some of the high end, but it's about the minimum value that I found cut off most of the noise without much loss in clarity. Just an idea.

Sibob · Post by **Sibob** » 11 Oct 2013, 10:07

Reporting back on the two SHO builds.
To my ears, the metal-film vs carbon film had minimum effect on the noise-floor, increased hiss present when pedal engaged.
The LED pop circuit did change the character of the pop, making it more of a click than a thump, which is benifical, I'll continue to play about with the circuit with varying cap values.
Again, I think the SHO circuit is perhaps not the best to experiment with LED popping as it seems inherient in the design (mentioned on ZVEX site), I've got some other clones to finish which I'll include it on.

Thanks for the help guys, practice makes perfect huh

In case you're interested, here are the two finished pedals with designs (that I didn't draw, but did cut) and stenciled on. They're relevent interests of the customers

Si

gjcamann · Post by **gjcamann** » 11 Oct 2013, 15:12

I would still recommend using a soldered star ground configuration. Without this, you won't have a reliable ground connection between your jacks. It may work, but if the aluminum on the box oxidizes a little (hint-it already has) or the nut becomes loose at all you'll loose your reliable ground. If you don't have a good ground connection, it will be like putting a resistor inline with the signal weather your pedal is on or off, and it'll sound better or worse as you jiggle it.

Also, did you try jymaze's suggestion

1) Against the " pop" : Add a resistor from the IN point to ground. Something between 2.2 and 4.7 meg should be fine in order not to load your pickups too much. If desperate, 1 meg will work but you may loose some sparkle.

I have a breadboarded SHO on the workbench that doesn't pop - and I always use the input resistor as described by jymaze. Take a look at some madbean pedal pdf's and you'll see that most (if not all) have this input resistor too.

galmar · Post by **galmar** » 11 Oct 2013, 17:10

I would still insist on following one of the 4 options I have previously stated. I think it is completely wrong to use the chassis as your ground. Think about the path that the output signal has to follow . And also read my previous post again.

Sibob · Post by **Sibob** » 11 Oct 2013, 23:47

Hi guys,

Sorry, I missed reading a few posts before my last update, didn't mean to seem like I ignored them!
Will test the grounding options above as well as the various cap & resistor placements for noise/pop.
Again, thanks for all the input, a lot to go through at this early stage in my building 'career', but intensely interesting

Cheers
Si