Bogner Helios Help: Killing Radio/RF

[FourT6and2]

I recently got a Bogner Helios 50. I like it. But there's a good bit of RF/radio noise. The amp is really two different amps in one. "Plex" input and "Hot" input are two different "amps". Not the traditional hi/lo jacks. Plex input bypasses V1. Plex input = no noise. Hot input = tons of noise: hum, buzz, radio stations. With no instrument cable plugged in = no noise. With cable plugged into Hot = noise. If I temporarily ground the instrument cable's barrel to the chassis at the jack = noise disappears. Perfectly quiet.

The jacks are Cliff-style. So they are isolated from the chassis. There is a 0.022uF cap from the Hot jack's ground, to chassis ground. So this should kill RF, theoretically. But the noise is there. The jack is also grounded via a separate wire to chassis ground. V1a input grid has no grid stopper. I was about to install one to attempt to kill the noise, when I tried grounding the cable and discovered that worked.

So... what's going on here? How can I solve this noise problem? The jack seems to be grounded already. And there's already a cap installed. Somewhere, there's a ground problem and I can't seem to track it down.

[FourT6and2]

Photos:

https://images-wixmp-ed30a86b8c4ca88...MI4BASTlCMzp4k

https://images-wixmp-ed30a86b8c4ca88...CS03zkQwsheeAs

https://images-wixmp-ed30a86b8c4ca88...bDbG19wlE9WSZg

https://images-wixmp-ed30a86b8c4ca88...3Q3IM6sJ-rIaEE

[g1]

When you say cable plugged in, it is connected to a guitar? Is the noise at all gain settings? Have you tried other tubes in V1 position?
Is there a schematic available?

[dai h.]

I think you already solved it (sort of). Lots of handwiring = lots of opportunities for error. I very coincidentally had a guts pic of a Helios, and it seems to show the V1 shield cable ground going to the ground lug (instead of the next one over as in your picture, which apparently(judging from your description) leaves the input ground floating). So, maybe try moving the "NOT GROUNDED" wire one over?

[FourT6and2]

When you say cable plugged in, it is connected to a guitar?

Doesn't matter. With or without a guitar plugged in, there is noise with cable plugged into amp. I've tried multiple cables, including the ones with the "silent" tip that disconnects when no guitar is present. So even though no guitar is plugged in, that side of the cable is shorted. But the side plugged into the amp breaks the jacks short and allows RF in. The instrument cable acts as an antennae, picking up radio. If I move the cable around, it changes the radio station lol. Without anything plugged in, the jacks are shorted to ground, so no noise.

With a guitar actually connected, there is still noise. It's not the guitar. It's not the cable. I have other amps here that are dead quiet.

Is the noise at all gain settings?

Yes. And no. Yes, in that the EQ/gain/switches don't affect noise. But no, because this amp's inputs are not typical hi/lo. The "Plex" input is a slightly different circuit and bypasses V1 and does not use the tone stack. The "Hot" input uses V1 and the entire tone stack. It's almost like having two different amps in one. Anyway... the radio and hum is on "Hot" input only.

Have you tried other tubes in V1 position?

All tubes have been swapped. Amp re-biased.

Is there a schematic available?

Here is one in-progress found online. It's not 100% and I haven't verified against actual amp.

[FourT6and2]

I think you already solved it (sort of). Lots of handwiring = lots of opportunities for error. I very coincidentally had a guts pic of a Helios, and it seems to show the V1 shield cable ground going to the ground lug (instead of the next one over as in your picture, which apparently(judging from your description) leaves the input ground floating). So, maybe try moving the "NOT GROUNDED" wire one over?

The shield connection at V1 shouldn't even be there. Normally the shield at the socket end of that wire is cut back and insulated under heat shrink. It should only be connected at one end (at the jack, to ground). If it's connected at both ends, it can't shield. It needs to be connected only at one end. I don't know why Bogner has it attached to a terminal strip. I took a photo of it because it's odd.

That said, yes... grounding the shield of that coax wire does reduce the RF/radio noise just like grounding the jack itself does because it's the same ground connection, electrically.

It's almost like Bogner left out a ground somewhere.

Here are some other photos I found online, it doesn't appear to be connected to ground in these photos either, so I think that connection is intended to be this way:

Connected to left-most lug on tag strip (not grounded):

[dai h.]

ahh... you're right. I seem to have mis-interpreted the picture (mine is fuzzier). Your point about the shield I agree with. Does seem a bit strange that the tube grid end is not just heat-shrinked off (what is the point of having the terminal strip??). I do still suspect that what you're describing with the weird noise is the 1M (or something) not being grounded or grounded correctly (also from the experienced improvement by connecting the cable ground (pickup return/shield) to chassis). If the input 1M ground side is open, then the pickup signal has some super high impedance path back to the pickup making it highly susceptible to noise (I would think).

[FourT6and2]

ahh... you're right. I seem to have mis-interpreted the picture (mine is fuzzier). Your point about the shield I agree with. Does seem a bit strange that the tube grid end is not just heat-shrinked off (what is the point of having the terminal strip??). I do still suspect that what you're describing with the weird noise is the 1M (or something) not being grounded or grounded correctly (also from the experienced improvement by connecting the cable ground (pickup return/shield) to chassis). If the input 1M ground side is open, then the pickup signal has some super high impedance path back to the pickup making it highly susceptible to noise (I would think).

Interesting. I might rewire the entire input side of the amp haha. I can do a better job but it's gonna be a pain.

[dai h.]

(You can call me lazy, but) I would (assess first, and) try to find an easy solution before you go "full bore" and start re-wiring everything. The green wires (which appear to be ground connections) from the jack to the underside of the board--are these going where they should, and are they connected (cold joints, etc.) correctly? In another one of the Helios chassis pics these are on the top side (I don't think it matters whether they are on top or bottom, but I can't see where they are going, and what apparent condition in your picture). Maybe do some more continuity and visual checking? The guitar cable shield ground has to go to jack, and the 1M(ground), then that should go to V1's 3k ohm cathode R(ground). Also losing the guitar cable shield to circuit ground connection would also result in serious weirdness (the super tiny signal current from the guitar pickup would return to the pickup, but through super high impedance paths). At this point (since it's Bogner) I'm presuming some sort of error in implementation rather than a serious design defect.

It's a little bit interesting how they seem to have gone with the traditional (as in same as older Marshalls) grounding scheme around the input with the ground bus across the pot backs and the insulated (Re'An?) jacks. I doubt there would be any sound difference with Switchcraft(non-insulated) but I suppose it would look less Marshall-like?

(almost forgot to post this "thought flash" lol... (Maybe a very long shot since the amp looks pretty new, but) could the jack contact (for the ground) be loose and not making proper contact with the jack ground?)

[Helmholtz]

If it's connected at both ends, it can't shield. It needs to be connected only at one end.

That's not true. As long as the shield is at ground potential it will be effective.
Otherwise your guitar cable wouldn't shield.

In an amp grounding both shield ends is typically avoided to minimize the risk of ground loops.
But if the shield only carries the signal return current there will be no problem - it's the normal use of a coax cable.

Is the amp new?
Have tried contacting Bogner?

[FourT6and2]

(You can call me lazy, but) I would (assess first, and) try to find an easy solution before you go "full bore" and start re-wiring everything. The green wires (which appear to be ground connections) from the jack to the underside of the board--are these going where they should, and are they connected (cold joints, etc.) correctly?

In another one of the Helios chassis pics these are on the top side (I don't think it matters whether they are on top or bottom, but I can't see where they are going, and what apparent condition in your picture). Maybe do some more continuity and visual checking? The guitar cable shield ground has to go to jack, and the 1M(ground), then that should go to V1's 3k ohm cathode R(ground).

Yep, those two green wires (one from Plex input and one from Hot) connect to the circuit ground on the PCB at V1 cathode and (-) side of V1 power supply cap. Look at the photo of the 100-watter I posted above. The entire V1 area of the PCB is a giant ground trace. All of those are connected to a PCB trace that is connected to the single terminal strip right next to them (where 0.022uF cap on input jack connects. And that terminal strip is screwed into the PCB standoff, which is chassis ground. ADDITIONALLY, the white wire connected to the same spot as the two green wires runs back to the bus bar soldered to the back of all the pots. I don't know why Bogner chose to run multiple connections to chassis like this. IMO, circuit ground should only have one connection to chassis ground.

I'll check the solder joints. But it'll be hard since it's a PCB and the connections are on the underside. It looks to be thru-plated/double sided. But there might be a clear coat on the top.

This is my ground scheme when I build an amp. Bus bar floating above pots, only connected to pot lugs that need it (not case backs), then input jack, then cathode of V1, then down the board to whatever needs it, then to filter caps, and then finally to one SINGLE chassis ground. My amp is dead quiet. Zero noise.

Also losing the guitar cable shield to circuit ground connection would also result in serious weirdness (the super tiny signal current from the guitar pickup would return to the pickup, but through super high impedance paths). At this point (since it's Bogner) I'm presuming some sort of error in implementation rather than a serious design defect.

Yeah the guitar cable shield seems to not be connected to circuit ground. But there are 1, 2, 3 redundant connections for the jacks to circuit ground and chassis ground. So I don't know what's going on unless one or more of those joints is bad. I'll check 'em.

It's a little bit interesting how they seem to have gone with the traditional (as in same as older Marshalls) grounding scheme around the input with the ground bus across the pot backs and the insulated (Re'An?) jacks. I doubt there would be any sound difference with Switchcraft(non-insulated) but I suppose it would look less Marshall-like?

All I know is it's ridiculous for a company like Bogner to wire an amp like this. The circuit itself sounds great. The amp's tone/timbre, etc... all great. It's just noisy as hell and has many redundant ground connections. That bus bar is grounded to the chassis like 10 times. Circuit ground has more than 8 connections to chassis.

(almost forgot to post this "thought flash" lol... (Maybe a very long shot since the amp looks pretty new, but) could the jack contact (for the ground) be loose and not making proper contact with the jack ground?)

It's possible. Only way for me to check that is to just replace the jack and I might as well rewire the inputs if I do that.

[dai h.]

I get that the modern(?) way of doing a grounding scheme has only one point for circuit ground to connect to chassis ground, but the older sort of way in Marshalls and Fenders of using the chassis for circuit ground connections can (and does) work. I don't think the "multiple connections to ground" for the pot backs is some sort of ground loop problem (as some seem to interpret it). To me, it's more of:

1) a "brute force" sort of approach to lower ground impedance ( = makes it more of a single connection where it's harder for a voltage drop to occur which isn't a problem because no high currents are handled by these returns)
2) the back-of-the-pot bus provides a convenient tie point for grounds, and
3) provides redundancy in case a pot nut comes loose ( = nothing happens but this wouldn't be the case if the redundant connections weren't there)

If you mentally fold the chassis front and back panels (and sides) back down to the flat piece it originally way and imagine the chassis as a ground plane (guess not as conductive as copper but it's pretty big so low impedance?), then maybe it makes a bit more sense. If multiple connections to ground didn't work, then the Helios preamp PCB ground plane area wouldn't work (although it doesn't sound like it's working as it should at the moment I doubt that is a problem) and ground planes wouldn't work (but placement of grounds should need to be correct and not just carried out in random fashion). (re: ground planes: In Dave Funk's book, he says (something like), "if it hums, it's a ground loop, and if it doesn't, it's a ground plane". Bruce (of Mission Amps in some very old post if memory serves) said (something like) imagine the ground point increasing in size. Also companies like Bogner, Suhr, THD might be using thicker foil like 105um (4oz?) or whatever to help lower ground impedance as well.)

Also, (from reading info about grounding op amp circuits) I wonder if there is some sort of "mojo"(some sort of coloration useful to a musician when not trying to create a linear(neutral) type of signal path) from doing it the old way (doing things the wrong way for op amp circuits can apparently result in more even harmonic distortion). Again, I would really try to figure it out instead of rushing in and re-building large parts of it (and maybe contact Bogner for assistance as Helmholtz said--maybe this is some sort of known problem that has occured in other units). It might a simple solution like a loose connection, iffy solder joint, a screw not tightened down enough as it was supposed to be to break through aluminum oxide on the surface or PCB resist or whatever. Anyway I'm not any sort of expert but I like reading about grounding which might be weird, but fun (for me anyway, lol...).

[Helmholtz]

BTW, how does the RF interference manifest? Can you hear a radiostation?

[FourT6and2]

I get that the modern(?) way of doing a grounding scheme has only one point for circuit ground to connect to chassis ground, but the older sort of way in Marshalls and Fenders of using the chassis for circuit ground connections can (and does) work. I don't think the "multiple connections to ground" for the pot backs is some sort of ground loop problem (as some seem to interpret it). To me, it's more of:

1) a "brute force" sort of approach to lower ground impedance ( = makes it more of a single connection where it's harder for a voltage drop to occur which isn't a problem because no high currents are handled by these returns)
2) the back-of-the-pot bus provides a convenient tie point for grounds, and
3) provides redundancy in case a pot nut comes loose ( = nothing happens but this wouldn't be the case if the redundant connections weren't there)

If you mentally fold the chassis front and back panels (and sides) back down to the flat piece it originally way and imagine the chassis as a ground plane (guess not as conductive as copper but it's pretty big so low impedance?), then maybe it makes a bit more sense. If multiple connections to ground didn't work, then the Helios preamp PCB ground plane area wouldn't work (although it doesn't sound like it's working as it should at the moment I doubt that is a problem) and ground planes wouldn't work (but placement of grounds should need to be correct and not just carried out in random fashion). (re: ground planes: In Dave Funk's book, he says (something like), "if it hums, it's a ground loop, and if it doesn't, it's a ground plane". Bruce (of Mission Amps in some very old post if memory serves) said (something like) imagine the ground point increasing in size. Also companies like Bogner, Suhr, THD might be using thicker foil like 105um (4oz?) or whatever to help lower ground impedance as well.)

Also, (from reading info about grounding op amp circuits) I wonder if there is some sort of "mojo"(some sort of coloration useful to a musician when not trying to create a linear(neutral) type of signal path) from doing it the old way (doing things the wrong way for op amp circuits can apparently result in more even harmonic distortion). Again, I would really try to figure it out instead of rushing in and re-building large parts of it (and maybe contact Bogner for assistance as Helmholtz said--maybe this is some sort of known problem that has occured in other units). It might a simple solution like a loose connection, iffy solder joint, a screw not tightened down enough as it was supposed to be to break through aluminum oxide on the surface or PCB resist or whatever. Anyway I'm not any sort of expert but I like reading about grounding which might be weird, but fun (for me anyway, lol...).

By multiple connections to ground, I don't mean like each pot casing. It's hard to explain without having the amp in front of you so I can literally point at parts of the circuit. But there are components that quite literally have two wires going to chassis ground lol. This amp is full of ground loops and redundant connections. And yes, I think grounding to the back of pots is ridiculous in a modern amp like a Bogner.

[FourT6and2]

BTW, how does the RF interference manifest? Can you hear a radiostation?

Hum, buzzing, and multiple overlapping radios stations.