research
 Patents
 Books
 Literature
 Articles
 Timeline
 Auction Archive

about
 About this site
 Wanted to buy

bulb gallery

Incandescent:
C
carbon
WD
drawn tungsten
WC
coiled tungsten
WM
mini tungsten
WS
pressed tung.
FG
figural bulbs
XL
christmas
XS
christmas sets
T
tantalum

Discharge:
NE
neon lamps
AR
argon lamps
XE
xenon lamps
MA
mercury
MC
fluorescent
MS
special mercury

Hardware:
F
fuses
FX
fixtures
PF
plugs & fittings
SA
sockets
SW
switches

tube gallery

 X-ray
 Geissler
 Crookes
 Radio
 Box art

museum pics

 Dr. Hugh Hicks
 
Fort Myers, FL.
 S.Slabyhoudek

links

 Related links
 Submit a link

 

Author Topic: Radio Frequency Interference from Incandescent Lamps?  (Read 21431 times)

Offline Chris Kocsis

  • Sr. Member
  • ***
  • Posts: 84
Radio Frequency Interference from Incandescent Lamps?
« on: September 15, 2008, 11:26:08 am »
Some collectors in the antique radio club I belong to, along with some collectors in Germany, are investigating a curious phenomenon of RFI in the FM band from squirrel-cage lamps operating above 160 volts.  The interference happens only above that voltage threshhold and cannot be duplicated with 110-volt lamps.  Does anyone, especially James and Chris and any others whose line voltage is 220 or 240, have any experience with such interference, or know of any documentation and theoretical explanation?  It would have been noticed probably no earlier than the very late 1930s.  I  can't find any references in my American literature.

Speaking of squirrel-cage lamps it seems like the only repros available now are Ferrowatt and a German brand Paul something.  Is anyone else making repros, especially for 220 or 235 volts?

All the best,

Chris
« Last Edit: September 15, 2008, 11:29:38 am by Chris Kocsis »

Offline Howard

  • Jr. Member
  • **
  • Posts: 35
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #1 on: September 15, 2008, 02:51:27 pm »
I have seen this demonstrated, once, at a lecture given to the Lothians' Radio Society, in Edinburgh, by the then Post Office Radio Interference Detection Department, c.1970.  The effect occurs when the lamp is powered with AC and is due apparently to the rapidly reversing Edison effect between the hot negative end of the tungsten filament and the cold(er) positive lead-in wire at the other.  This produced broadband RF hash, modulated by 100Hz (double the supply frequency - one burst for every half cycle) across the HF and VHF bands.

Sadly, I cannot reproduce this with the one 240v squirrel cage filament lamp in my collection.  Therefore, a question:  were some of these lamps gas filled?  The Edison effect only occurs in a vacuum, so only vacuum lamps would demonstrate the effect.

My guess as to why the effect cannot be produced with 110v lamps is that the spacing between the two lead-in wires has something to do with it, also a shorter length of filament significantly more negative than the opposite lead-in wire.
« Last Edit: September 15, 2008, 02:55:22 pm by Howard »
--
Regards, Howard.
"Is there any tea on this ship?" - Arthur Dent

Offline Chris Kocsis

  • Sr. Member
  • ***
  • Posts: 84
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #2 on: September 15, 2008, 04:16:32 pm »
I hadn't realized that it happens only in vacuum lamps.  They made it happen with a modern German repro -- so I'm a little surprised it isn't gas-filled.

Offline Howard

  • Jr. Member
  • **
  • Posts: 35
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #3 on: September 16, 2008, 05:36:37 pm »
Oops!  I left out an important part of my previous explanation.  Nearly forty years have dulled my memory...

To produce the amount of RFI which is produced requires a resonant circuit.  This is the inductance and self-capacitance of the squirrel cage filament itself.

It may be of interest that the interference was demonstrated as a 100Hz buzz at 100Hz intervals across much of (read the entire coverage range of the communications receiver used!) HF and VHF.  On a VHF monochrome TV, the interference was manifest as two broad horizontal bands of "snow".  At this demonstration, these were stationary, showing that the local mains supply was synchronised with that at the TV studio, to which their timebase generators were locked.  Any variation would cause the noise bands to move slowly up or down.

...

Chris,  I'm still puzzling over the gas filled or not thing.  I haven't got any ideas yet.
--
Regards, Howard.
"Is there any tea on this ship?" - Arthur Dent

Offline Max

  • Sr. Member
  • ***
  • Posts: 72
  • 100W cleartop mercury lamp during run-up
    • the Lamp Review
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #4 on: September 18, 2008, 05:24:44 pm »
Hi,

I do not have a direct experience with this effect, however I have read in an old German Osram publication that this effect occurs in lamps whose vacuum is imperfect. What happens is that the electric field present between each strech of filament drives the electrons thermionically emitted from the white-hot tungsten (i.e. the Edison effect, which does occur in gases as well). When the field is strong enough and there is a presence of a gas at low pressure, the latter will become ionized and will suddenly enable a stronger current to flow between filament sections. Since the E-field fluctuates with the applied voltage across the filament, the discharge will strike and extinguish itself during each AC half-cycle, leading to transient current peaks responsible for the emission of EM interferences.
The reason why this effect occurs preferably at higher main voltages is simply because the field strength between filament sections is stronger, therefore increasing the chance of ionizing any residual gas present inside the lamp. This EMI emission vanishes if the vacuum in the lamp is high enough.

Max

Offline Mónico González

  • Hero Member
  • ****
  • Posts: 146
  • Philips HP-80w Mercury Lamp
    • Mis Bombillas, peque?o museo virtual de la l?mpara el?ctrica.
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #5 on: October 08, 2008, 08:40:00 pm »
Hi,
I?ve observed this effect too, more exactly on Paulmann's modern replica's like this from my collection:

http://mis-bombillas.com/Mis-Bombi-Web/Rustika_MF_230_60.htm

This phenomenon is very noticeable at 100 MHz (middle of VHF-FM broadcast band), where an intense 100 Hz buzz can be heard when a receiver tuned at these frequencies are placed at a couple of metres from the lamp or the power line to which are connected it.
Obviously this has to do with thermionic emission (Edison effect), because one end of the long filament acts as a cathode during a half cycle, precisely these that is negative, while the other one being positive acts as anode, just like the plate in a diode. But a simple diode powered either with DC or AC cannot generate any self-maintained oscillation because a diode isn't an amplification device. Also as have been said, to get any oscillation an RC oscillating circuit is needed, as well as any way to periodically transferring energy to it from the power circuit.
The explanation must be searched for the special structure of these kind of lamps in wich, some parts of the filament could act as the "grid" of a triode. As the whole filament are subjet to voltage variations, a kind of biasing could result of this. Also this biasing are modulating in a very complex mode the whole structure from which could result a true amplification phenomenon.
I think that the fact of fails in vacuum has nothing to do with this RF generation, because all of us know that the amplification process in "gassy" tubes are most of times greatly blocked by gas ions, so, a high vacuum inside the bulb could eventually help to the amplification-oscillation phenomenon.
Cage lamps are, of course, high vacuum bulbs, so I guess this could have to do with RF emission of this kind of lamps more than an eventual low pressure gas traces.
The frecuency range of emitted energy, are undoubtely fixed by the relationship between stray inductance and capacitance of filament structure also the filament lenght. The fact that lamps rated for higher voltages are more prone to generate RF, probably has to do with the higher acceleration of electrons in a more intense E-field due to a greater potential between instantaneous cathode-anode sections of filament.
As the "buzz" can be tuned either in AM and FM modes, I guess the "carrier" generated by the lamp are modulated in a very complex way in both modes. However I haven't found any evidence of interference in the hectometric (medium waves) AM band coming from these lamps, even using tube receivers, more sensitive for E-field component of the wave.
Regards.
« Last Edit: October 08, 2008, 08:44:34 pm by M?nico Gonz?lez »

Offline Chris Kocsis

  • Sr. Member
  • ***
  • Posts: 84
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #6 on: October 09, 2008, 04:54:54 pm »
What makes the experiment by the radio collectors (which started this thread) interesting is that they are using pure DC, not AC.  As pure, at least, as a Lambda supply can make it.  In a few weeks I will try to summarize the experiment here (I'm waiting to see if there are more comments from the radio community to research I reported in the first post, which was mentioned in an article in the newsletter of the Mid Atlantic Antique Radio Club).  But in the meantime I have a limited number of copies of the newsletter containing that article which I could mail to interested readers.  I can't reproduce the article here in full because it was already offered to Lud Sibley for the Tube Collector journal.   

Cheers,

Chris
« Last Edit: October 09, 2008, 05:25:54 pm by Chris Kocsis »

Offline James

  • Hero Member
  • ****
  • Posts: 169
    • www.lamptech.co.uk
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #7 on: October 26, 2008, 05:50:11 pm »
Hi Chris,

This is certainly a known effect in the high voltage squirrel cage lamps.  I recall producing them at GE where we identified that the interference increased as a function of poor vacuum quality.  When the lamps were exhausted to vacuum better than 10^-3 mmHg pressure there was no discernable interference.  However poorer levels of vacuum trigger a current flow in the residual gas between the two ends of the filament across the lead wires, which leads to the RF emissions.

The problem could be reduced or even eliminated by spraying the filament mount assemblies with a cryolite-rich phosphorus mixture so as to attain higher vacuum in production.  The lamps were lit on a 300V supply immediately after being removed from the exhaust machine, with a ballast in series, and as the cryolite-phosphorus mixture is vaporised it is ionised and the bulb fills with a brilliant blue glow.  Within about 2-3 seconds the glow collapses and the getter functions with high efficacy to pull a hard vacuum in the bulb.  Providing this is done, and the glass is heated to within 25 degrees celsius of its softening temperature during exhaust, then you get a good vacuum and no interference.  However it's challenging to run to such limits, and normally the vacuum quality is somewhat inferior.  Consequently from batch to batch you often find lamps which produce the interference.  Sadly EU regulations on radio interference mean that these lamps are no longer legal for sale here, because the interference prevents them from being CE marked.

NB these lamp types are always vacuum!  If they are gas filled you get a wonderful effect where the hot gases rise to the top end of the bulb due to convection currents.  The lower area of the filament emits no light and only the upper end, surrounded by the warm gases, is incandescent.  Upon inverting the bulb, the gases recirculate and the other end lights up!  I have made samples of these for training lectures on lamp technology, photo attached to show the effect.  Vacuum lamp on the LHS, and Gasfilled with 85:15 Argon:Nitrogen at 700mmHg on the RHS.  Note also how much dimmer is the gas filled lamp, due to heat losses to the gas filling.  It demonstrates very nicely the principles of Langmuir's discovery of reducing thermal losses from the filament by coiling.
« Last Edit: October 26, 2008, 05:53:54 pm by James »

Offline Chris Kocsis

  • Sr. Member
  • ***
  • Posts: 84
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #8 on: October 26, 2008, 10:24:21 pm »
Hi James, thanks for the reply and the neat pictures.  I didn't know that even the higher wattages of squirrel-cage lamps contained a vacuum.

But what do you say to the finding, in the experiment reported by the radio group, that the RFI occurred from bulbs operated on DC, not AC?

I'm waiting to see if these experimenters report anything further before I try to summarize the article. 

Everyone who wants a hard copy of the article, please email me.

Chris 

Offline Mónico González

  • Hero Member
  • ****
  • Posts: 146
  • Philips HP-80w Mercury Lamp
    • Mis Bombillas, peque?o museo virtual de la l?mpara el?ctrica.
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #9 on: October 30, 2008, 10:04:18 am »
Curiously, lamps as which you can watch at my collection:

http://mis-bombillas.com/Mis-Bombi-Web/Rustika_MF_230_60.htm

shows the typical red-powdered spray at the glass shaft that means that these lamps have been passed by the phosphor gettering stage during exausting. However, this one as other of its tungsten cage "companions" in my collection, shows the annoying RF emission phenomenon, although its vacuum degree seems to be high enough.
But these lamps are yet CE labelled, marketed and easily found even owing the strict communication laws on Europe.
The fact that a high vacuum squirrel lamps could be capable of generating RF energy I think is irrelevant about the kind of current it could be powered with. In any case, with pure DC we will get also pure or not modulated RF instead the strongly "hum" modulateted RF radiated from an AC powered bulb. Furthermore, if we use a DC source with superimposed audiofrequency, we could make, at least theorethically, a true low power radio transmitter.
Regards.

Offline hygroman

  • New Member!
  • *
  • Posts: 1
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #10 on: January 04, 2009, 07:51:36 pm »
I must preface this post by saying that I am not a collector. I happend on this site while searching for information on RFI from incandescents wondering if any one else had observed the phenomenon. I noticed it in the 1950"s whille doing RFI testing inside a screened room. One day while testing a piece of military hardware the noise level suddenly increased dramatically all across the spectrum we could measure with the equipment we had at the time. There was nothing I knew of in the room that could have caused it so at first I suspected the receiver. After illiminating that and any break in the integrity of the shielding, I finally tracted the source to one of the 100 watt incandescent lamps we use to light the room. The lamp was still lit but was putting out noise like crazy. After unscrewing the bulb the noise disappeared. I replaced the bulb and got back to work thinking no more of it until it happened again about six months later. My theory at the time was that the filament was broken but arking across the tiny gap, the plasma created allowing the current to flow thereby keeping the lamp lit.The explanation you folks have for the squirrrel cage lamps is quite interesting but do you think it is the same cause as in the bulbs I observed which were operatiing on the standard 117 volt US supply. Incidently, I reported the phenomen to the military back then in my reports but then forgot all about it.

Bob Q.

Offline Chris Kocsis

  • Sr. Member
  • ***
  • Posts: 84
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #11 on: January 06, 2009, 04:31:37 pm »
Hi Bob, arcing was definitely not the cause in these experiments.  If you (or anyone else) would like a photocopy of the latest information that was published in my antique radio club's newsletter, give me your snail mail address and I'll mail it to you.

Chris

Offline andrewdeandrade

  • New Member!
  • *
  • Posts: 1
Re: Radio Frequency Interference from Incandescent Lamps?
« Reply #12 on: December 11, 2019, 03:05:19 pm »
So, I'm not sure if anyone will see this thread since it's been over 10 years since it was last commented in, but I came across this exact RFI issue with Edison bulbs running at 110V mains. Basically, I was setting up my terminal and getting a strong buzz through my tonearm. After much trial and error, I isolated the problem to some Edison bulbs.

I'm going to include some links to videos I took.

This video shows the hum mostly disappearing when I turn off the squirrel cage Edison bulbs:

https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FAaR5xrfZ_w8P8d7XqFPxdTfpJe6YARTD7sfNVA193HXnDonqa9VvWIzu%2F%24%7Bf%7D%3Fo%3DArS2MW5F5jyDgM7MZ0UijbckA7qFfmEy4YLOi5fzg0tF%26v%3D1%26x%3D3%26a%3DCAogwRhIMwlYchIsff68mIJC4a7NjEPXrX-ETGcAdx0f_4MSJxDussWc7y0Y7sLA8PgtIgEAKggByAD_CEgpeFIE6SXumFoEb1iM7g%26e%3D1578636419%26k%3D%24%7Buk%7D%26fl%3D%26r%3D828C5B5E-B509-49DA-8E9E-A3F3963AFD96-1%26ckc%3Dcom.apple.largeattachment%26ckz%3DFE5EBBA6-773B-41AB-8D30-A5F523FE5BA2%26p%3D52%26s%3D5kkD9-Aco5Pvb60hQBE9l8JpqfI&uk=EyvrKQT00me00uJaAk4ywQ&f=Video_1.mov&sz=1086505

This video shows the hum amplified when I touch the light fixture in which the bulbs are installed:

https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FAVi3Gb2n-Q0i6i51oLenhF_IQfWPAU5jF_jx_W4hvK55I_qlQ7OJ1kgp%2F%24%7Bf%7D%3Fo%3DAiNFnC91KL5daYrkfEUwzfOjCrmuvBVtQmN4KXoYYmdA%26v%3D1%26x%3D3%26a%3DCAog5O40VHdO_YuPtAfxfb8w52ENHgzyov2klU0LgE3IoV0SJxCKtcWc7y0YisXA8PgtIgEAKggByAD_IeJ631IEyEH1j1oEidZIKQ%26e%3D1578636419%26k%3D%24%7Buk%7D%26fl%3D%26r%3DC0893FC6-F83C-49BD-A236-943690D0BCCA-1%26ckc%3Dcom.apple.largeattachment%26ckz%3DFE5EBBA6-773B-41AB-8D30-A5F523FE5BA2%26p%3D52%26s%3Dii-PG9t7PVv8dTN7bZxz_0LCM8A&uk=mgQiqcBtFSp0T96KbrxiCQ&f=Video.mov&sz=1905515

Here's a video I took where I place some aluminum foil over the tonearm. When the foil touches the large aluminum platter, it appears to help act as an RFI sink and reduces the amount of RFI in the tonearm

https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FAVrtGWYTA_c2IsryI-8mhH9xZMbPAdOIgf2g2Fg_hCCjuNEaX016k-ya%2F%24%7Bf%7D%3Fo%3DAleEVFVAYjof6bbArWaf50VR3JQJPc0nFJAi3xJZ87-U%26v%3D1%26x%3D3%26a%3DCAogNIkMBLywGqPhrb1z6akc5mR-8Ev5l8sQB5gC5znv01ASJxDVt8Wc7y0Y1cfA8PgtIgEAKggByAD_SAcNe1IEcWTGz1oEepPsmg%26e%3D1578636420%26k%3D%24%7Buk%7D%26fl%3D%26r%3D7470E5F0-F3D8-4452-B16D-3F24DD91842D-1%26ckc%3Dcom.apple.largeattachment%26ckz%3DFE5EBBA6-773B-41AB-8D30-A5F523FE5BA2%26p%3D52%26s%3D3g7LTQaShEK3r3zfo-RdJIRTqnQ&uk=lpYhGbxijbW5Ts4vSdZl2Q&f=Video_2.mov&sz=10473049

Here are a few other videos of experiments to try and figure out what was causing the RFI:

https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FAdTufidMtk9W_ZsVgN9eiZIqHnQLAR9NNBIPbeqoLV89NadoN10sbNqX%2F%24%7Bf%7D%3Fo%3DAr0vQGb2n2Opcn809US58QYua2R9E_7lZiGxoCunlbhh%26v%3D1%26x%3D3%26a%3DCAogBSfD87HAJzK0Xn84WWkZnmspCcluCDrndlxjTGi13IwSJxD2s8Wc7y0Y9sPA8PgtIgEAKggByAD_ERO8U1IEKh50C1oELGzalw%26e%3D1578636419%26k%3D%24%7Buk%7D%26fl%3D%26r%3D63D3FFFD-C39E-4782-9EFC-7D4BE13058FE-1%26ckc%3Dcom.apple.largeattachment%26ckz%3DFE5EBBA6-773B-41AB-8D30-A5F523FE5BA2%26p%3D52%26s%3DTHlQtK1ce_yPzNRCICHXMCkQo14&uk=KbCoNgzdjd-V_UXObJinzQ&f=Video_3.mov&sz=5278028

https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FASoVGBi3FTH-CvEvcoX2dx_MjTt1AUUiSTQ7senxr_i06yS2QDaneeQ8%2F%24%7Bf%7D%3Fo%3DAquKsU5rG4UlICfIZWtQEYEqzyBOYKQ-9fnicPVSxRbS%26v%3D1%26x%3D3%26a%3DCAogfLzy1KSpLmCbBVHIrRqKQILEWHen-ZpLU6Xpn7tFhzQSJxDmtsWc7y0Y5sbA8PgtIgEAKggByAD_WbznX1IEzI07dVoEp3nkPA%26e%3D1578636419%26k%3D%24%7Buk%7D%26fl%3D%26r%3D56C3C916-4650-4889-8424-230482F7E4B5-1%26ckc%3Dcom.apple.largeattachment%26ckz%3DFE5EBBA6-773B-41AB-8D30-A5F523FE5BA2%26p%3D52%26s%3DcpLm5ilOahlVnwn5N4jQfOVxAk4&uk=-HPB8fsvrYJ_rU-uGL61xw&f=Video_4.mov&sz=15021729

https://www.icloud.com/attachment/?u=https%3A%2F%2Fcvws.icloud-content.com%2FB%2FAWJiLLWZxE9gvuQmBT-HA--oBvreAc55PK8HTwFV40ex-tcqdahacnHP%2F%24%7Bf%7D%3Fo%3DAkTGhcOwjmIRzdebke_BA_4dugDPQ1nnQ0C0iQEICNUh%26v%3D1%26x%3D3%26a%3DCAogKwhXQDQJ43Ausep95VelzGctxkF1IMv_hPkdRE_00PsSJxCCtsWc7y0YgsbA8PgtIgEAKggByAD_ekK281IEqAb63loEWnJxzw%26e%3D1578636419%26k%3D%24%7Buk%7D%26fl%3D%26r%3DEDCE10D7-CBDC-441C-9188-2CFBF62093F7-1%26ckc%3Dcom.apple.largeattachment%26ckz%3DFE5EBBA6-773B-41AB-8D30-A5F523FE5BA2%26p%3D52%26s%3DlNhgsgkMw61-WXmMCsVs6-Kp7YI&uk=8CqKaS20BxOPY21iygf3EQ&f=Video_5.mov&sz=4233650

If these links those work, I'll see if I can upload them elsewhere. Unfortunately. *.mov files can't be uploaded here.

In the spirit of contributing to the knowledge here, the bulbs have no markings on them except 120V/60W etched in the brass bass. They are most likely brand new bulbs bought by the homeowner (I rent) and the have a filament that zig zags 12 times (there is a six-pronged filament support at the end of the bulb). This is the most similar one I could find online: https://www.colorcord.com/products/edison-vintage-light-bulb?variant=12363324719206

I don't know if there is any gas in them or if they are a vacuum, but I would guess that these have a vacuum as they are cheap modern reproductions.

If anyone wants to do any experimentation here let me know. If anyone knows if there is a way to solve the RFI without replacing the bulbs let me know. If the only option is to replace the bulbs, what kinds of bulbs maintain the same look and feel but don't have an RFI problem. If I need to replace them, I'm happy to send them to anyone who might want one. I have 5.