RFI issue

[TubeGeek]

I built an 18W amp for a guy that lives near an FM transmitting tower and of course he is getting strong rf through his amp.


The input signal wiring is shielded and there are 33k grid stoppers on pins 2 and 7 of the 12ax7.


I have applied shielding tape on the corners of the chassis to try to block any rf from leaking into the chassis, shielded tape on the back panel of the cabinet too.


I have installed a 0.01uF cap from the "ground" side of the input jack to the chassis, this reduced the rfi but did not fully block it.


Any other tricks you guys can share? 

[eleventeen]

It will to some degree change the character of the amp if you change it permanently. but maybe you have to go lower gain with a 5751 or 12AY7 for the first tube. I'd at least experiment with such.


I have no idea where the best place to put one (or two) but perhaps an RF choke in series with V1 grid or in series with (a reduced size) plate resistor on V1 might help.

[vibrolax]

I have installed a 0.01uF cap from the "ground" side of the input jack to the chassis, this reduced the rfi but did not fully block it.

The input jack ground is DC-isolated from the chassis?  That is, ring terminal of the input jack gets its DC ground from the circuit board, and the AC ground to the chassis via the 0.01 uF cap.  Ceramic cap?  I don't know if a larger cap might help.

In my trainwreck express-like amp, I used a 0.001 uF ceramic cap ring terminal AC ground to chassis and 10K grid to V1, and that was good enough for my situation.  Sounds like you have the basics covered.

Jon

[shooter]

Is the *bottom* side of the chassis Open?, I had pretty bad RFI from my invertor, cut an aluminum plate and copper taped it to the bottom and knocked the RF way in the dirt

[PRR]

> a 0.01uF cap from the "ground" side of the input jack to the chassis

How physically big is the cap? 100MHz waves are kinda short.

I think in the FM band, 0.001uFd (1nFd, 1,000pFd) with VERY short leads is more likely to shunt waves.

Shields on low-level tubes? We all hate these, but under a tower they may be needed.

Demodulating FM with g-amp type circuitry is very odd.

[shooter]

Demodulating FM

fwiw, both my Son in IN. and me in MI have picked up radio stations, but only when plugged in, both stations were low power Christian stations, Neither of us could find the station with a *real* radio   Suspect AM though, not FM

[TubeGeek]

The transmitting frequency is 100.1 MHz, CBC radio in Canada.


The input jacks are UK cliff style and the ground from the jacks is connected to a preamp buss bar, which is connected to the chassis near the jacks.  I drilled a hole and connected the buss bar with a ring terminal.


Something I just thought of is that I grounded the shield of the rg-174 coax to the chassis at the v1 socket, not the preamp buss bar.  Maybe grounding the shield at the input jacks would be wiser?  It's something I am willing to try if the rfi would be better eliminated with this method. 


I used a ceramic cap and did not focus on lead length.  Good point.  I'll try and come up with a way of mounting it with shortest as possible leads.


Using a 5751 or 12AT7 is another option, I'll try it out.


The bottom of the chassis is open and I have covered the section where it is open on the cabinet board with aluminum duct tape to help with shielding the open chassis.  The corners of the aluminum chassis aren't welded so I also duct taped them up.


I have also installed shields on the preamp tubes.


This has been a problem with the owner ever since he has lived in this house.  He has had city engineers, CBC tech's, electrician's, other amp tech's to his house troubleshoot.  It comes through his TV's and appliances too.  He has tried ferrite beads that were provided to him by the radio company to help block it.  It's a pretty big pain in the butt for him.  I have agreed that I'll make a list of things to try and I will go to his house and do them, cause I don't get the issue where I live, which is about 4 or 5 km's away.


This may be a bit of a challenge to resolve.  He needs a faraday caged house.

[HotBluePlates]

This has been a problem with the owner ever since he has lived in this house.  He has had city engineers, CBC tech's, electrician's, other amp tech's to his house troubleshoot.  It comes through his TV's and appliances too.  He has tried ferrite beads that were provided to him by the radio company to help block it.  It's a pretty big pain in the butt for him.  I have agreed that I'll make a list of things to try and I will go to his house and do them, cause I don't get the issue where I live, which is about 4 or 5 km's away.

Well, if all those other folks can't help him, I doubt you can do much either.

I kind of wondered why you didn't raise the input grid stopper from 33kΩ to the typical 68kΩ or higher.

Then again, even if you shield everything at the amp, then you have the cable to wonder about. And if you get the cable squared away, you have to wonder about his guitar. And once the guitar is squared away, maybe the pickups still pull in RF interference.

The only good answer is get away from the transmitter.

[mresistor]

I was checking a Twin Reverb out at a music store and had taken off the back panel. I have a MotoG Gen 3 and I had the led flashlight function on and was looking at the plates of the 6L6's. Whn I got the phone up close to the tubes it generated a nasty noise through the amp. Maybe shield the power tubes like the 9 pins are shielded. 

[kagliostro]

Just to try, install (in addition to the shielded cable, resistors and capacitor) a ferrite bead innmediately before the grid stoppers resistors

(Something like in Mesa amps that are without the grid stopper and uses the ferrite bead)

---

Can you place an aluminium sheet as shield on the open side of the chassis ?

Franco

[TubeGeek]

This has been a problem with the owner ever since he has lived in this house.  He has had city engineers, CBC tech's, electrician's, other amp tech's to his house troubleshoot.  It comes through his TV's and appliances too.  He has tried ferrite beads that were provided to him by the radio company to help block it.  It's a pretty big pain in the butt for him.  I have agreed that I'll make a list of things to try and I will go to his house and do them, cause I don't get the issue where I live, which is about 4 or 5 km's away.

Well, if all those other folks can't help him, I doubt you can do much either.

I kind of wondered why you didn't raise the input grid stopper from 33kΩ to the typical 68kΩ or higher.

Then again, even if you shield everything at the amp, then you have the cable to wonder about. And if you get the cable squared away, you have to wonder about his guitar. And once the guitar is squared away, maybe the pickups still pull in RF interference.

The only good answer is get away from the transmitter.

I was able to help him with his Allesandro amplifier last year.  It would be nice to be the guy that solves the problem but I agree with you, getting away from the transmitter would be best.


Since I built him his 18W, I have to do my best to block the rf if at all possible.


I did recommend that he have his guitar's shielded as well and he did have it done just recently.

[TubeGeek]

Just to try, install (in addition to the shielded cable, resistors and capacitor) a ferrite bead innmediately before the grid stoppers resistors

(Something like in Mesa amps that are without the grid stopper and uses the ferrite bead)

---

Can you place an aluminium sheet as shield on the open side of the chassis ?

Franco

Hmmm, I didn't realize they made ferrite beads small enough for this.  I'll give this a try if I can find some.

[Willabe]

I did recommend that he have his guitar's shielded as well and he did have it done just recently.

Did it help any?

[TubeGeek]

I did recommend that he have his guitar's shielded as well and he did have it done just recently.

Did it help any?

He said it didn't help.  The amp has radio while idling so it has to be the amp.  I'll be going over to his place next week with a list of things to try.

[sluckey]

Something I just thought of is that I grounded the shield of the rg-174 coax to the chassis at the v1 socket, not the preamp buss bar.  Maybe grounding the shield at the input jacks would be wiser?  It's something I am willing to try if the rfi would be better eliminated with this method. 

Connecting shields at the source end is the way it's done in facilities such as an Air Traffic Control tower. The input jacks would be your source end. Will it make a difference if you live under a transmitter antenna? Probably not.

The bottom of the chassis is open and I have covered the section where it is open on the cabinet board with aluminum duct tape to help with shielding the open chassis.  The corners of the aluminum chassis aren't welded so I also duct taped them up.

Does that aluminum tape make good electrical contact with the chassis? It needs to.

This has been a problem with the owner ever since he has lived in this house.  He has had city engineers, CBC tech's, electrician's, other amp tech's to his house troubleshoot.  It comes through his TV's and appliances too.  He has tried ferrite beads that were provided to him by the radio company to help block it.  It's a pretty big pain in the butt for him.  I have agreed that I'll make a list of things to try and I will go to his house and do them, cause I don't get the issue where I live, which is about 4 or 5 km's away.

Don't beat yourself up over things you can't control.

[TubeGeek]

Something I just thought of is that I grounded the shield of the rg-174 coax to the chassis at the v1 socket, not the preamp buss bar.  Maybe grounding the shield at the input jacks would be wiser?  It's something I am willing to try if the rfi would be better eliminated with this method. 

Connecting shields at the source end is the way it's done in facilities such as an Air Traffic Control tower. The input jacks would be your source end. Will it make a difference if you live under a transmitter antenna? Probably not.

The bottom of the chassis is open and I have covered the section where it is open on the cabinet board with aluminum duct tape to help with shielding the open chassis.  The corners of the aluminum chassis aren't welded so I also duct taped them up.

Does that aluminum tape make good electrical contact with the chassis? It needs to.

This has been a problem with the owner ever since he has lived in this house.  He has had city engineers, CBC tech's, electrician's, other amp tech's to his house troubleshoot.  It comes through his TV's and appliances too.  He has tried ferrite beads that were provided to him by the radio company to help block it.  It's a pretty big pain in the butt for him.  I have agreed that I'll make a list of things to try and I will go to his house and do them, cause I don't get the issue where I live, which is about 4 or 5 km's away.

Don't beat yourself up over things you can't control.

Yes I made sure the aluminum tape does make a sealed connection with the open end of the chassis.


I won't fret too much over this if I try what I can.

[kagliostro]

I didn't realize they made ferrite beads small enough for this

You can find ferrite beads also in small format





http://www.budgetronics.eu/index.php?action=article&aid=19066&lang=EN





Franco

[Mike_J]

Is there any negative tonewise to placing a ferrite bead over an RF174 cable from the input jack to the grid of the preamp tube? They are not very expensive.


Thanks
Mike

[kagliostro]

I think that the part of wire that is to be put across the ferrite bead must not be shielded


Franco

[HotBluePlates]

Is there any negative tonewise to placing a ferrite bead over an RF174 cable from the input jack to the grid of the preamp tube? ...

A grid stop resistor works against a tube's Miller capacitance to roll off extreme highs. A 68kΩ resistor against an average 110pF input capacitance is -3dB at ~21kHz (above hearing).

A ferrite bead increases the inductance of a piece of wire (more so if the wire can be looped through the bead several times, as in Kagliostro's picture). Inductance is a rising impedance as frequency rises. But the bead usually turn the wire into a fairly-small inductor. Even a bead which bring the wire inductance up to 100mH only looks like ~12.6kΩ at 20kHz.

So we're not talking big impacts in the audio range; it would take a lot more series resistance and/or inductance to do that. FWIW, your amp's speaker very likely has serious roll-off by 6kHz, and your guitar is harmonics-only at 1kHz & beyond.

[El Rollo]

The transmitting frequency is 100.1 MHz, CBC radio in Canada.


The input jacks are UK cliff style and the ground from the jacks is connected to a preamp buss bar, which is connected to the chassis near the jacks.  I drilled a hole and connected the buss bar with a ring terminal.


Something I just thought of is that I grounded the shield of the rg-174 coax to the chassis at the v1 socket, not the preamp buss bar.  Maybe grounding the shield at the input jacks would be wiser?  It's something I am willing to try if the rfi would be better eliminated with this method. 


I used a ceramic cap and did not focus on lead length.  Good point.  I'll try and come up with a way of mounting it with shortest as possible leads.


Using a 5751 or 12AT7 is another option, I'll try it out.


The bottom of the chassis is open and I have covered the section where it is open on the cabinet board with aluminum duct tape to help with shielding the open chassis.  The corners of the aluminum chassis aren't welded so I also duct taped them up.


I have also installed shields on the preamp tubes.


This has been a problem with the owner ever since he has lived in this house.  He has had city engineers, CBC tech's, electrician's, other amp tech's to his house troubleshoot.  It comes through his TV's and appliances too.  He has tried ferrite beads that were provided to him by the radio company to help block it.  It's a pretty big pain in the butt for him.  I have agreed that I'll make a list of things to try and I will go to his house and do them, cause I don't get the issue where I live, which is about 4 or 5 km's away.


This may be a bit of a challenge to resolve.  He needs a faraday caged house.

One thing to also check is the wiring in the amp any wire close to the resonate frequency will act like an antenna and will supply a stronger RF signal into the amps circuit. Ferrite beads of the right mix might help alot(yes there are many types of ferrite mix's) Check some Ferrite makers websites to see what would be the best mix for blocking the VHF signal the radio station. My guess is that some wiring needs to be changed.   

[TubeGeek]

Unfortunately nothing has helped reduce the rf interference.


I spent a couple hours there this morning and had a list of 9 things to try.


I am in an awkward position now because he feels he has a custom amp that he cannot enjoy.  He asked if I could refund his money, which is something I didn't expect to have to deal with.

[eleventeen]

IMO, your best bet is to do so and instigate forward movement on your part. Simple answer, I know.

[HotBluePlates]

Unfortunately nothing has helped reduce the rf interference. ... I spent a couple hours there this morning and had a list of 9 things to try.

I am in an awkward position now because he feels he has a custom amp that he cannot enjoy.  He asked if I could refund his money ...

I had a feeling it would go this way back when I said that if others couldn't help him, you probably wouldn't be able to either. I was worried he'd blame his location problem on your build...

IMO, your best bet is to do so and instigate forward movement on your part. Simple answer, I know.

... but I do agree with Eleventeen. Now you're here, and he's asking for a refund. It's not your fault. But for good karma and word-of-mouth reputation, you should probably take the amp back and try to find it a new home. Hopefully, being an 18w, it will sell quickly.

[TubeGeek]

We agreed to talk about the outcome in a few days.  It gives us both some time think over what to do.


He is not upset at all but he feels stuck.  I get his situation.  I live a few blocks from him and it the amp sounds good at my place. I did mention to him when I agreed to build the amp that I cannot guarantee no rf in the amp.  I said I would do my best to eliminate it but if I couldn't, that it was not my fault.


He is thinking a solid state amp may not have the rf issue.  I didn't feel like explaining more.


What stings is another little detail when the amp was built....the US to Canada exchange rate was at 49% plus don't forget international shipping costs as well.  Ugh, oh well.  I will do what is proper and move forward.  It is not worth the stress and annoyance to have bad word of mouth or karma.  A bonus is I get a really cool amp back in my collection! I really did like how this one came out with it having a new celestion cream alnico speaker installed.

[eleventeen]

Not much you can do. In business, I have always worked on the principle that either we have a happy deal or we take it apart and restore where we were beforehand. That has served me very well and I don't recall ever having permanently lost a customer thereby. I may have lost the subject deal. So be it. Obviously, you wanted to execute the build and collect the $$, and he wanted the amp, or there never would have been the initial agreement to build it. Undoing the deal, *cheerfully* tells the client that you value your work product as much as you were willing to sell it to him for. You'd both be happier with him having the amp and you having the $$ but sometimes the best intentions simply do not work out, on nobody's fault. You gave it a serious shot, that's all you can do. Undo the deal, move on.

[John]

Good advice. Doesn't make it sting any less, I know.


And now you know at least one kind of job to turn down. ;)

[alerich]

I will do what is proper and move forward.  It is not worth the stress and annoyance to have bad word of mouth or karma.

Good call and good for you. I have a feeling that whatever costs you incur in this transaction will pay dividends down the road in your amp business. In all fairness, it does sound like you had an inkling of what you might be getting into all caveats aside. No more custom amps for this chap.

I feel for him. It would suck to find a home that you otherwise enjoy that makes it impossible to enjoy your hobby. I live in a 100 year old farmhouse that doesn't have the quietest power in the world but it is tolerable. The upside is that I live in a 100 year old farmhouse with no nearby neighbors and I have loud hobbies. A little AC noise now and again is no sweat.

Back in my citizen band radio days my paltry 4 watt rig always messed with the TV signals of my neighbors (this was late seventies largely before the cable TV boom) and it was always my fault. Hard to believe that this radio station owner isn't held to a higher liability level. They should pay for the amp.

[Ritchie200]

+1 take it back and don't look back.  When the next one does the same thing, nothing bad will be said.

Demodulating FM

fwiw, both my Son in IN. and me in MI have picked up radio stations, but only when plugged in, both stations were low power Christian stations, Neither of us could find the station with a *real* radio   Suspect AM though, not FM

Are you sure they were local low power stations?  There are several Christian shortwave broadcasts on several bands.  There are four big ones blazing at 100,000 watts each at New Orleans, Nashville, Knoxville, and Memphis.  I can sometimes pick them up with a combination of vintage pedals.

Jim

[Ritchie200]

Back in my citizen band radio days my paltry 4 watt rig always messed with the TV signals of my neighbors (this was late seventies largely before the cable TV boom) and it was always my fault. Hard to believe that this radio station owner isn't held to a higher liability level. They should pay for the amp.

CB 11 meter band was exactly half of the VHF channels 2-6 in MHz.  So, any overmodulation (especially), and antenna design and placement will show on your neighbors TV.  Think tube amp harmonics for an analogy.  Many of the value priced CB's fulfilled the bare minimum FCC requirements for interference and were well known for "splatter".  Some of the better ones like Browning, Robyn, Teaberry, Johnson, etc, and the old crystal controlled radios were much better. 

I'm sure the radio station is well within it's FCC regulations because adjacent stations (in frequency) would cry foul if they weren't.  Living under an antenna is a bummer as I'm sure any steps for attenuation are just being overpowered.  Kind of like when the bar you are playing at has the stage next to all the neon beer lights.

Jim

[trobbins]

Did you rf bypass the speaker leads at their chassis connector as well?

It may help to add in some first stage grid to spigot/0V rf bypass capacitance.  C shouldn't be too high to affect audio band, but should try to shunt away as much as possible through shortest lowest impedance path - as per the bypassing of the input socket.

[drgonzonm]

I found a website, that provides some suggestions on addressing RFI, "alectrosystems.com/tips/audio_tip_rfi"
One thing this site suggests is balanced signals using xlr plugs. 

If you need tube solutions to converting unbalanced signals to balanced signals and back. brodski at tubecad.com has some designs. 

I did not see, any discussion, that the guy with the RFI problem is using other (newer design) amps.  I am under the assumption that he is. 

From what I've read, you have converted the amp to a faraday box, so the signal pickup may not be internal to the amp.  Hence, the balance signal suggestion. 

[mat janssen]

Put a ceramic capacitor value 47pF from inputjack to gnd on each channel, on each plug.
Regards M

[shooter]

Are you sure they were local low power stations

Nope, you're probably correct since I tried a couple times *running the radio dial* and not finding it.

my paltry 4 watt rig always messed with the TV signals

I worked fixing Cb's in the '70s, the owner had a base unit, 1000watt *box* and a moonraker 4 antenna, every time he key'd up you might as well go on break

[TubeGeek]

Update:


The owner and I spoke and he does not want to return the amp to me after all.  He is happy with what I built him.  He has had rf issues with every amp he has had in his house, it is what it is when you live next to a transmitting radio tower.  When he takes the amp out to a buddies, the amp is good. 


I'll be willing to continue investigating solutions when time permits.  He is a repeat customer and I'm happy to tweak the amp here and there.  I am in the stages of slowing down my project intake to break for summer.  I like to get to the mountains as often as possible in the summer months.


One thing I do recall is that the rf interference was most strong when I touched the preamp ground buss with my needle nose pliers.   

[DummyLoad]

One thing I do recall is that the rf interference was most strong when I touched the preamp ground buss with my needle nose pliers.

have you tried installing a 51R MF 1/4W and .01uF in series from preamp ground to chassis?

with this scheme you have two ground points: one is physically tied to the chassis, the second (preferably the ground end of input jack) is tied to the RC series network that's grounded at a different point on the chassis.

--pete

[shooter]

I like to get to the mountains as often as possible

+1, tho my climbing days are over, the mountains still call