Stopping buzz in my homemade spring reverb

[GrahamGrant2]

I have my homemade spring reverb set up and it’s sounding pretty decent, but there’s terrible hum / buzz coming from it.

The setup is mono aux out of desk into 9volt aby pedal to duplicate the signal. These two jack cables then go to a sub zero hum destroyer box and then from jack to rca cables to two reverb tanks. The rca outs from the two tanks go into jack inputs on another hum destroyer box and then trs out to xlr inputs on the desk. I noticed by accident that if I have one hand touching both tanks at the same time and then the other hand touching two of the barrels of the two output jacks (going back to the desk) at the second hum destroyer the noise completely goes away.

So…given my electrical knowledge is minimal and I expect a kicking from the experts…can I connect an earth wire from one tank to the other and then another one from one barrel of the jack to the other? Or is this totally unsafe? Or do I just have magic hands?!!

[James Perrett]

I suspect the hum destroyers are causing your reverb tank not to be earthed which means that the screening around the tank is less effective. Try removing both hum destroyers. If there is a hum problem then try using just one hum destroyer - try with one on the input and then one on the output and see which configuration is quietest.

Your wire idea should work but I don't like seeing things being made more complicated than they need to be.

[resistorman]

Why do you have the hum destroyer boxes hooked up at all? They're likely the cause due to the way they work.

[Wonks]

You should certainly only have the hum destroyer connected on the inputs or the outputs, not both. They break the ground connection via transformers, and if they are on both input and output, there is no ground reference for the tanks or the cables to and from the tanks to the hum destroyers.

Check the tank connections to the casing. You may find the case is only connected to just one of the connector grounds - which if so, means that you don’t need the hum destroyers. Or the case may need a separate ground connection (not common, but possible).

The output transformers in the tank are very prone to picking up hum - 12k ohms impedance means a lot of turns to boost the output - so need to be well away from any transformers and PSUs. Like a single coil guitar pickup, rotate the tank(s) until you get the least amount of hum.

If the hum destroyers are like the similar Behringer HD400, their own casing isn’t grounded which can allow the isolating transformers to pick up noise, especially if the casing was touched. I modded my HD400 to connect the casing to the ground connection of the left input, which made it really quiet.

[GrahamGrant2]

Thanks all - very helpful indeed. I think I ended up using two by mistake as I was looking to extend the cables to get the tanks away from my rack which was causing noise. Will play around today and see if I can minimise it, but in general will it be ok to add ground wires between components in situations like this if they fix the noise? Thank you!

[GrahamGrant2]

You should certainly only have the hum destroyer connected on the inputs or the outputs, not both. They break the ground connection via transformers, and if they are on both input and output, there is no ground reference for the tanks or the cables to and from the tanks to the hum destroyers.

Check the tank connections to the casing. You may find the case is only connected to just one of the connector grounds - which if so, means that you don’t need the hum destroyers. Or the case may need a separate ground connection (not common, but possible).

The output transformers in the tank are very prone to picking up hum - 12k ohms impedance means a lot of turns to boost the output - so need to be well away from any transformers and PSUs. Like a single coil guitar pickup, rotate the tank(s) until you get the least amount of hum.

If the hum destroyers are like the similar Behringer HD400, their own casing isn’t grounded which can allow the isolating transformers to pick up noise, especially if the casing was touched. I modded my HD400 to connect the casing to the ground connection of the left input, which made it really quiet.

Yes, it’s a behringer HD400 on the output. How did you do your fix? Solder a wire to the case and then to the cable itself, presumably by taking the cable apart?

Struggling to find a good place to get started learning properly audio electronics!

[Wonks]

Only if you keep both hum destroyers, otherwise you are adding in parallel ground paths and introducing possible ground loop issues.

For the HD4000 I scraped the paint off the top and bottom casing joins, so I got continuity for the casing. Then I soldered a small eyelet to the end of some wire and soldered the other end of the wire to the input 1 jack ground pin, then sandwiched the washer between the two halves where one of the casing screws went.

[James Perrett]

Struggling to find a good place to get started learning properly audio electronics!

The problem is that there is an awful lot of theory involved if you really want to learn electronics which appears to bear little relationship to audio. And the electronic engineer's name for some things is different to the audio engineer's name for it so you may find something like differential signalling described in detail in an electronics course yet miss the fact that audio engineers tend to call them balanced lines.

A great book for those who really want to take a deep dive into electronics is The Art of Electronics by Horowitz and Hill. I bought myself a copy a couple of years ago and really wish I had bought it sooner.

Douglas Self has written a few books on audio electronics which are interesting and may be easier to read. I have a copy of Self on Audio which is more hifi oriented while Small Signal Audio Design may be more applicable to what we do here.

Of course, Sound On Sound covers many technical subjects so it is worth digging around on the website. There is also a Useful Information Archive on this forum at

https://www.soundonsound.com/forum/viewforum.php?f=35

which includes information about grounding and solving ground loops.l

[Wonks]

There's a new version (4th revision) of Douglas Self's Small Signal Audio Design due out on 20th Nov 2023.

[Folderol]

There's a new version (4th revision) of Douglas Self's Small Signal Audio Design due out on 20th Nov 2023.

Interesting. I wonder what's been added/removed

[Hugh Robjohns]

For the third edition he added completely new chapters on tape machine electronics and guitar amps. Plus a load of updates and expansions throughout.

For the 4th edition it looks like he's adding content on internally balanced audio circuitry, electret microphones, emitter-follower stability, and microphony in capacitors. The last two subjects are pretty esoteric, the first is very rare, and many many people make their own mics? I

I'm sure it will all be fascinating and educational, but if audio electronics are of interest I'd buy the 3rd edition now rather than wait for a 4th edition in November. There's more than enough meaty stuff in the 3rd edition to keep anyone going for years....

[MarkOne]

A great book for those who really want to take a deep dive into electronics is The Art of Electronics by Horowitz and Hill.

Gosh! I wonder what happened to mine?

I had mine in 1980 when I was doing electrickery properly, designing and testing HiFi amps (this is before the loonies took over)

[Hugh Robjohns]

Small world! I bought my copy in 1980 too. Still refer to it, although I keep meaning to buy a more modern version.

[GrahamGrant2]

Thank you everyone for the very useful suggestions. I ordered Practical Audio Electronics by Kevin Robinson which arrived today. And as for the spring reverb, after an almighty strop on my behalf, it turns out it works much better if I just connect it straight to the desk. I think I was trying to convert the unbalanced outs into balanced ins on the desk but it just caused problems. It’s a bit quiet so need a fair bit of gain, but otherwise sounds lovely. I’ve been looking at diy plate reverbs recently….

[resistorman]

I’ve been looking at diy plate reverbs recently….

good on ya!

[GrahamGrant2]

I'm back to the spring reverb I'm afraid! I ordered some RCA (from tanks) to XLR (mixer) leads for it following a suggestion on a previous thread to use the mic input possibly to get extra gain. However, there's still hum on the system which is bugging me, so on a whim I decided to connect the XLR end of the RCA cable into the UNBALANCED XLR input of LD02 DI box which arrived today for reamping pedals and then connecting the BALANCED XLR output to the mixer. All of a sudden, not only has the hum gone, but the tone has totally changed - previously it has been a very dark reverb, like a dark plate, but now it's bright, sproingy and sounding, well, like how a spring reverb should. How can this have happened? I know certain configurations of cables can cause tone to change, but this is hugely dramatic. Confused, but also delighted!

[merlyn]

The reverb tank must be high impedance, and needs a high impedance input. If you don't have a high impedance input, treble frequencies are lost. If it's because the reverb tank is inductive, then the treble frequencies never make it out of the tank. If it's because the tank is capacitive, then the treble frequencies are shorted to ground.

[GrahamGrant2]

Thank you. Whichever of those three it is, the DI box has managed to salvage them! Although, without it, it does have a nice very dark plate quality albeit too noisy.

[Wonks]

With the tank's output impedance of 12k ohms, you'd need around 60k ohm minimum input impedance to avoid loading the output and loosing treble, but higher would be better.

An active DI box should have a high enough input impedance, but a passive one plugged into a mic input might only give 30k or so. Better than going straight into a line input with a 10k impedance, but higher is better here.

This won't really affect noise, that's all going to be down to grounding and shielding and keeping the tank away from strong noise sources.

[Folderol]

This won't really affect noise, that's all going to be down to grounding and shielding and keeping the tank away from strong noise sources.

Often overlooked: the tank will pick up mechanical noise as well as electrical.

[merlyn]

Thank you. Whichever of those three it is, the DI box has managed to salvage them!

There has been debate elsewhere as to how many yards out of nine to go ... but I think we can go a little further.

It's only two things. Saying 'the impedance is 12k' is probably actually saying the DC resistance is 12k. Impedance varies with frequency. The combination of the reverb tank and low input impedance has formed a low pass filter, and that can be done with a capacitor, or an inductor. If you're interested you could look at filters. You don't have to though, and the rule of thumb is that the nominal impedance of an input should be ten times the impedance of the preceding output.