Hum problems with GEO ABY box....any ideas?

[Paul Marossy]

The transformer that is NOT being used is left with its primary open. That lets the secondary act like a pickup coil for magnetic fields. If you could short the unused primary, it should stop.

My money is on that being the case.  

[bryantabuteau]

cool.  that makes sense to me.  I'll try that after work today.

[Gilles C]

...This may have to do with there being a strong magnetic field near where you're doing testing,, or just the hum gods in general....

Hey, that sounds good. I happen to have strong magnetic field somewhere in my house (lot of wires in the walls close by??), so it's a good environment to test effects...  :roll:

For example, I have to orient my guitar correctly when I play if I don't want to hear any hum with my guitar plugged straigth into the amp....

I'll wait for the results. It could decide me to look for mine.  :wink:

GillesC

[Mike Burgundy]

Sounds very good - but this still bugs me: Why is it Always the Twin?
That's slightly weird, on first glance...

[bryantabuteau]

i would guess that because of the resistors i put in that tied the outputs to the internal ground, the isolation got shot, and the normal ground difference hum is occuring.  So when i go and try this other method - right after this post actually, i'll try it first without the resistors.

[Gilles C]

I have a Fender Hot Rod Deluxe, and I get hum when it is not selected

Gilles C

[bryantabuteau]

maybe its the fender   Nah,  I just did some rewiring,
and it has eliminated the hum in either amp when on A or B.  makes sense.  I am now grounding the non-grounded input of the primary of the unused output isolation transformer when each is switched.  I have removed the resistors that were added earlier on in the thread.  

Now, the one problem remaining.  That of rethinking the switching.  One that i couldn't solve with a quick think.  When you select the Y switch, it bridges the two inputs of both isolation transformers, one of which is now tied to ground, so of course, no output on either.  Great as a mute switch, but not that good as a Y switch.   So the switch has to tie the two inputs and LIFT the ground on the grounded input.    

maybe someone has some expertise in switching.  its confusing me.

[bryantabuteau]

I can work this out using 2x 3dpdt switches, but not using the 2x 2dpdts that are currently in there even without LEDs.   Anyone with smart ideas?

[R.G.]

Hey, great! Nailed the hum.

I'll think about the switching a bit. Maybe there's a trick or two we can use to do the LEDs and grounding.

F'rinstance - change to shunt instead of series switching. Use a single input buffer, feeds two secondary buffers, one on each transformer primary. Each secondary buffer is fed through a 10K resistor from the output of the primary buffer. The A/B switch has its pole grounded, and each throw goes to the input of a secondary buffer. Now the switch does not let signal through the desired channel, it grounds the un-desired channel. The /Y switch un-grounds the pole of the A/B switch, so neither is grounded.

You do the same with the LEDs. The pole of the LED side of the A/B switch is connected to ground. Each A and B LED has a resistor from +9 to the LED, which has its cathode grounded. Each throw of the A/B LED switch is connected to the resistor/LED junction on one side. Then whichever channel is muted has its LED turned off by being shunted to ground. The same trick works for the /Y LED - it un-grounds the path from the LED pole to ground, and both LEDs come on.

Shunt switching requires thinking sideways compared to series switching.

[bryantabuteau]

Interesting.  Looks like I'll have to do some reading to understand all that  I think I get it, but am the sort of person that needs to see diagrams to understand concepts.  I'll go read up on shunt switching.  What sort of thing are you envisioning for the buffers?  what would they consist of?

[bryantabuteau]

I was about to try typing it all out, but visually is much better,

whats wrong with this diagram.  you lose the Y LED, but evrything else might work

http://bgmt.2dhost.net/switching.gif

switch on left is a/b switch on right is Y

[R.G.]

Here's what I meant:

http://geofex.com/FX_images/humfree2.gif

[Ed Rembold]

Maybe I'm missing something, but couldn't isolation be achieved
with just 1 transformer coupled output, and the "other" output coupled normally through the box.
Ed R.

[Ed Rembold]

Maybe I'm missing something, but couldn't isolation be achieved
with just 1 transformer coupled output, and the "other" output coupled normally through the box.
Ed R.

[bryantabuteau]

I'll leave that question to someone qualified to answer it.  I'm just building from a schematic that someone made up since i don't have the skills yet to design my own.  in any case, heres my modified switching diagram for use with 3dpdt switches to fit with the old schematic.  The previous one i posted was soooo wrong it wasn't funny.   This hopefully means i can put the new switches in, rewire and it works.   At some point I'll try a rebuild with buffers once I understand the concepts invloved, but its on perfboard now and to make those sort of changes to the circuit would require me to start from scratch.  So, anything clearly wrong with this wiring setup?

http://bgmt.2dhost.net/3dpdtswitching.gif

[R.G.]

Maybe I'm missing something, but couldn't isolation be achieved
with just 1 transformer coupled output, and the "other" output coupled normally through the box.

Yep, it could. That would leave the input always connected to one amp's ground, and that probably would never cause any problems.

I used two because
(a) any tone effects from the buffers and transformers are sent equally to both outputs
and
(b) the circuit is the son of the transformer coupled splitter.

Actually, the right way to do all this is to have the **amp** have a diffamp input with balanced inputs and high common mode rejection. The amp would then be effectively blind to common mode hum.  And it would use one more triode section than we use now. Sigh.

[Ed Rembold]

Thanks R.G.
By the way, I loved your "switching solution", beautifully clever.

Ed R.

[Ed Rembold]

Thanks R.G.
By the way, I loved your "switching solution", beautifully clever.

Ed R.

Edit- sorry guys I don't know why I'm double posting.

[bryantabuteau]

Cool, I think i understand that diagram now,  had a bit of a think about it.   Now what sort of buffers would you be using?  just the same as the input buffer?  So i could socket up another LM833 and use it for the two output buffers?

[R.G.]

Now what sort of buffers would you be using? just the same as the input buffer? So i could socket up another LM833 and use it for the two output buffers?

Yep, another 833 buffer will work fine.

By the way, I loved your "switching solution", beautifully clever.

Aw, shucks, t'weren't nothin...

Shunt switching is hard to get a handle on at first, but once you do, it's really handy to be able to flip between series and shunt switching. A little bit like the mind change needed to go from thinking about what the voltage is doing in a circuit over to where the current is flowing.