| Pages:
1
2 |
densest
Hazard to Others
Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline
Mood: slowly warming to strain point
|
|
@watson.fawkes - Agreed! Some of the meters look to be more useful & competently designed: decade attenuators & wide-band (500MHz?-5GHz?)
response. Your point about testing with a reference signal is a good one. I don't think any of them would survive immersion in a full power field.
http://www.repairfaq.org/REPAIR/F_micfaq6.html has some notes on DIY leakage testers - old technology (neon bulb) and new technology (LED). The
LED version could be made much more sensitive by using a low capacitance Schottky diode - probably a 100V one or even ($$) a SiC diode.
http://www.mdpi.com/1420-3049/6/10/831/pdf has a diagram of a modified oven used to synthesize phthalocyanine pigments.
There is a lot of material published over the last 15 years on using modified microwave ovens for chemistry. The most common way to shield holes in
the oven is to extend a copper tube or mesh out from the oven at least 10cm with the conductive shield bonded to the inside of the oven. I don't have
access to most of the journals. Somewhere in that pile of information must be a discussion of how to engineer such a shield.
One very interesting thread in the literature is whether non-heat-related effects occur. The consensus I saw was that careful evaluation of the
products showed hot spots or areas in all cases. Still, one wonders about the high speed nitrations described some months ago on this web site - there
doesn't seem to be any place where hot spots would occur, especially not enough to do bulk reactions.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by densest  | I don't think any of them would survive immersion in a full power field.
[...]
One very interesting thread in the literature is whether non-heat-related effects occur. |
That full power
field, it's a bitch. And you really want a testing device that can withstand it. Suppose, by way of an example use, that you have an exit pipe in the
wall of the oven; it's a common example. Assume that the cross-section has dimensions of less than λ/4 so that you're down below the
waveguide cutoff. (Note to reader: if this makes no sense to you, don't modify your microwave; you don't know enough. Start by orienting yourself with
the Wikipedia article on electromagnetic waveguides and then start studying.) You could just check the output leakage, but what you really want to do is to sample the
attenuation in your pipe by sticking your probe down the tube at various depths and ensure it's attenuating correctly. You're in a field with a
strength far above leakage, and it's good to have a device that can not be destroyed in the process. (It's OK if it saturates and doesn't give you a
reading, but you don't want it to fail permanently with this kind of test.) If this seems unnatural, note that it's generic about any significant
leakage that comes from a design or fabrication flaw. It's exactly when you have such a flaw that you want your test equipment to tell you.
Public Safety Announcement: Note that microwave ovens operate at a frequency designed to heat water and that your eyes are full of water. Please do
not cook your eyeballs. See the repairfaq section on microwave ovens for non-standard applications.
As for the second point about non-heat-related effects, I would say that there aren't any, but that there are effects related to an out-of-equilibrium
heat distribution, where some bonds are hotter than others and the chemical reaction time is faster than the heat relaxation time. This is partly
semantics, since "heat" usually means equilibrium heat or steady-state heat flow. This situation is one where you have one bulk temperature and then,
at least for a short time, a higher temperature for particular bonds. I'd say there are going to be some interesting pathway selection effects.
|
|
|
densest
Hazard to Others
Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline
Mood: slowly warming to strain point
|
|
| Quote: Originally posted by watson.fawkes | That full power field, it's a bitch. And you really want a testing device that can withstand it ... and it's good to have a device that can not be
destroyed in the process.
|
A meter or so of RG-8 cable, a small tuned loop, a 50-ohm noninductive load with a series capacitor to allow DC continuity testing of the load, an
electrically isolated thermocouple with RF choke(s) in the leads, and a temperature meter - total outlay near $0 if one has the thermocouple &
meter, $30 or so for a cheap multimeter with thermocouple connections.
Ideally, some shielded twisted pair - maybe 2 pairs out of a high performance Category 6 shielded Ethernet cable in parallel, something magic from a
new 10Gb copper Ethernet, DisplayPort, etc. would be better. Maybe some old FiberChannel copper twisted pair? But RG-8 coax with the 9.5AWG center
conductor would work and withstand a lot of abuse.
A 1-5W incandescent bulb (auto brake light or turn signal) would be really tough to burn out and even vaguely the right impedance. One of the no-base
wire lead ones would be good though it would require scraping the leads to bare metal to solder it. It is even an indicator of truly high power levels
Localized nonuniform energy distributions? Wouldn't a 2.5Ghz resonance be a molecular one or a group of molecules? Dunno...
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
If, by a "tuned loop", you
mean one with a resonant frequency at 2.4 GHz, then that antenna won't fit down an output pipe that's supposed to be attenuating. The cross-sectional
dimension of the pipe is, by design, smaller than (say) λ/4, but the resonant dimension is at λ/2. This is a bit of an
over-simplification, but makes the essential point that such an antenna won't do everything you want. This device, however, would be reasonable for
doing certain kinds of high-strength work, but you don't want to make it hand-held. Maybe on the end of a very long pole. A proverbial ten-foot pole.
I spoke with my friend again about this subject. He pointed out that the analogue of a proper field strength meter is a motorcycle helmet, that is, a
modest buy-in cost of a fun and dangerous activity. Verbum sapienti satis est; a word to the wise is sufficient. All these low-end devices
are useful, but at this point I'm of the opinion that they are secondary measurement devices, not primary ones.
A few other notes. The antenna in your meter may be polarized. Remember to rotate the antenna to check for field strengths at different polarizations.
Cracks in the conductive envelope of the oven will radiate (and the radiation will be polarized). The rule is that every possible dimension needs to
be below an appropriate length limit. This is one of the reasons that waveguides are bolted together with lots of bolts around the perimeter. These
bolts are acting as conductors to keep the radiation inside. For example, if you decide to make a demountable pipe with threads and teflon tape, it
will leak radiation out, and quite a lot, because the pipe is now electrically isolated and acts as a passive re-radiator. So all joints should be
hard-soldered (or brazed or welded), if permanent, or use lots of metal bolts at close spacing, if demountable.
The wavelength at 2.4 GHz is 12.5 cm, just a bit less than 5 inches. So it's reasonably to think about 1" ID pipe as an exhaust pipe, but it has to be
long enough. Using 1/2" pipe would be better, because it attenuates faster. Fortunately, these dimensions are large enough that standard dimension
glass tubes will fit inside them without any particular hassle.
Ensure that a gaseous output product is not a conductor at microwave frequencies. If it is, then it acts just like a coax cable in the output pipe and
you'll need to do even more. Basically it means that the condenser has to go inside the shielded cavity, so that the output product becomes
non-conductive before it leaves the conductive envelope.
The inverse square law is your friend. Even a kilowatt of power from a point source generates small fields at enough distance. So take your
experimental apparatus out in a field with a long extension cord and approach it slowly, spiraling inward. Do not lean over it to test it from the
top; that's not approaching it slowly from that direction. Instead, turn it on its side and do another round of field strength testing. Scan the field
from your feet to your head while you're doing this.
|
|
|
densest
Hazard to Others
Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline
Mood: slowly warming to strain point
|
|
Agreed, again... do the engineering and then do a reality check! Finding out that corrosive gases chewed a hole in the shielding could ruin one's day,
so check every time. The advice about reliable & continuous shielding is well taken.
My concept of the search loop was a small (1-2 cm) coil which is either self-resonant or resonant with a small added capacitor (a few pF) reminiscent
of an old grid-dip meter. Yes it will be directional - one would have to rotate it or have several probes with different orientations. It will also
not register in a standing wave null. RG-8 is quite stiff and could be used as a pole. It's about 1.1 cm thick so it would fit through a useful sized
pipe.
Readings from anything amateur-built are of course a bit suspect  but any reading
beyond a very small one should be alarming! A simple probe which could detect (say) 10-100 mw/cm2 to check for gross problems (on a pole!) and a more
sensitive meter to poke around would be a lot easier to make than a real lab instrument with wide input range.
The literature I've seen so far says that thermocouples give the widest dynamic range and reliability for the fields we would be interested in. I've
seen a couple of "real" meters using thermocouple probes.
It would be instructive to take apart a few commercial field strength meters.
|
|
|
| Pages:
1
2 |
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
