Ionized Mercury / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Ionized Mercury

1 2 next

Move Thread

LAN_403

Myke

Mon Sept 28 2009, 11:03PM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

I was wondering if it is completely safe too look at mercury rectifiers while they are on. I can't seem to find any information saying that it is unsafe to look at the discharge but I am worried because ionized mercury gives off some rather short wavelengths of light. Does anyone know what sort of glass they use to make these mercury rectifiers and if it blocks UV?

If it isn't safe to look at, does anyone know of some cheap material that can block short wavelengths of light? Acrylic doesn't seem to block much UV because there is a coating they put on the surface to make it block UV.

Proud Mary

Mon Sept 28 2009, 11:48PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Mercury arc rectifiers can and do emit abundant UV - including short wave length UV.

The best form of UV shielding is an aluminium screening can.

klugesmith

Tue Sept 29 2009, 12:47AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

I've often been told to avoid looking at Hg-vapor rectifiers while they are on, because their glass envelope isn't intended to block UV light. But like you, haven't found an authoritative reference.
Need to find out, by measurement if necessary, whether UV-absorbing glass or plastic from a picture framing store (or window tinting) can make a shield for safely viewing the rectifiers. Then can finish a semiconductor-free MWO. Here is a proof-of-concept trial:

employing an old-style 866 (no stupid can around the filament) to help heat a mug of coffee.
[edit] here are some 866A's displayed front-and-center in some audiophile gear. Followed by a couple of comments that it's not a very wise idea.

See also

where breakage of a UV-absorbing outer envelope exposes the quartz arc tube.

aonomus

Tue Sept 29 2009, 04:16AM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

In general, the heavier the element, the more electrons in higher orbitals to promote and release energy on collapsing down to its original place, thus the more different wavelengths of light possible (broader spectrum). Also in general, the heavier the element, the greater chance of the energy levels being high enough for UV to be emitted. You can try using UV leds to see if the glass fluoresces under UV, but rather safe than sorry, just use a solid piece of metal in between you and the UV.

In one of the prep rooms at the uni, a petri dish machine uses a UV light to sterilize the plates/agar, and they just cover the plastic shield with aluminum foil to give it some more protection. Just be careful because your stuff will involve HV, and the foil can act like a capacitor if its in the wrong place.

VulcanBB18

Tue Sept 29 2009, 04:57AM

Registered Member #2376 Joined: Mon Sept 21 2009, 05:13AM
Location: Brisbane, Australia
Posts: 14

From what I understand, normal glass blocks about 90% of UV rays.

A wikipedia (ugh)

source says : "Ordinary window glass passes about 90% of the light above 350 nm, but blocks over 90% of the light below 300 nm."

So wouldn't that mean the envelope would at least afford you some protection?

Proud Mary

Tue Sept 29 2009, 04:51PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

The 866 data sheet recommends shielding i both to protect the valve itself from interfering EM fields, and to block RFI, of which mercury arc rectifiers are a prolific source.

Shielding may also be used to stabilize the bulb temperature, which is very important in the correct operation of the evaporation-condensation cycle.

Myke

Wed Sept 30 2009, 12:12AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

Thanks for all the info!
I wouldn't want to rely on glass to block sub 350nm UV because it would let UV around 400nm though pretty much unaffected.
For the shield, would I have to blow air though it so it doesn't get too hot? How much UV is produced compared to visible light? Would the EM energy get higher as the tube passes more current?

Proud Mary

Wed Sept 30 2009, 12:54AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Myke wrote ...

Thanks for all the info!
I wouldn't want to rely on glass to block sub 350nm UV because it would let UV around 400nm though pretty much unaffected.
For the shield, would I have to blow air though it so it doesn't get too hot? How much UV is produced compared to visible light? Would the EM energy get higher as the tube passes more current?

Hi Mike,
remember that 866 was marketed in 1938, when no one would have thought it possible that someone would consider setting up an HV circuit without a metal case.

As for cooling, or warming, there is no need if you use it at room temperature. The 25 deg - 60 deg C temperature figures are all in the data sheet. Don't put it in a tight screening can - one of the early type with large holes cut all over it and a press-down lid will be fine, or a home made one using wire gauze/mesh. Yes, RFI from the tube will increase in rough proportion to the tube current.

Hytron 866

RCA 866

RCA 866A

Myke

Wed Sept 30 2009, 03:09AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

I have a 872a and it seems to get pretty warm. Is there something I could be doing wrong to cause it to heat up so much? I measured the filament voltage and it seems to be fine.

I was meaning "does the UV emitted shorten in wavelength as the current increases." I was thinking this becuase it would seem like the electrons would be falling a lot farther since they have more energy. The electrons have more energy as the current increases, right?

klugesmith

Wed Sept 30 2009, 06:41AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

I was meaning "does the UV emitted shorten in wavelength as the current increases." I was thinking this becuase it would seem like the electrons would be falling a lot farther since they have more energy. The electrons have more energy as the current increases, right?

Nope, the forward voltage drop (and electron energy) hardly increases (in a rectifier tube). You just get more carriers in the same volume. Mercury vapor LAMPS have various pressures, voltages, and spectral weighting, but I'm no expert.

I have learned a lot more about the eye safety issues.

Looks like any UV exposure hazard from Hg-vapor rectifiers would be from the 253.7-nm "germicidal lamp" spectral line. Can cause "welder's flash". The rays don't penetrate the cornea, much less get focused by eye. So it depends on exposure time and distance. Looking away from the source doesn't help, if the source can see your eyes.

Regular skin is shielded by a layer of dead cells at the surface. At this point I strongly suspect that emissions of UVA and UVB wavelengths, and of 185 nm, are of less concern.

Threshold Limit Values for UV exposure are given (as recommendations) by the ACGIH:

For UVA (400 to 315 nm) the limit is 1 mW/cm^2 if exposure time exceeds 1000 seconds, and 1000 mJ/cm^2 total dose for any shorter exposure.
For UVB (315 to 280) and UVC (280 down to less than 200 nm) it's strongly wavelength-dependent, and most conservative at 270 nm (3.1 mJ/cm^2 total in 8 hours).
The TLV for 254 nm is 6 mJ/cm^2, which is 0.2 uW/cm^2 for 8 hours, or illumination of EPROMs in an eraser (remember those?) for 1/2 to one second.
Table 1 in this international reference gives very similar numbers.

For me, the next step is to find the UVC brightness of the Hg-vapor rectifier glow. If 100 times less than a germicidal or EEPROM-erasing lamp, and observer were 10x farther away, then 1 hour would be OK but all day could burn your unshielded eyes. That's a lot of ifs in a radiation safety calculation.

A window of UV-absorbing acrylic would almost certainly make it safe.

Window glass would probably do, but I haven't yet found any data that is quantitative about both wavelength and thickness. Remember, if "plain window glass" absorbs 15/16 of some wavelength, the same material 4 times thinner (like a vacuum tube envelope) absorbs only 1/2.

Where are the experts who already know this stuff?
Rich

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob