"Transverse-ElectroMagnetic" - a waveguide term describing the orientation of the E and H fields relative to the direction of propagation (also intimating that their are *other* orientation modes e.g. TE or TM etc.). Jim ----- Original Message ----- From: "Lawrence Lile" To: Sent: Tuesday, May 28, 2002 8:33 AM Subject: Re: [EE]: Sensitivity training for Engineers > Ja! This is cumbersome, will take some days to build, some cash, but if > it is well done, you can also perform other tests in there. > As signal coupling device, I recommend a TEM cell, some kind of rhombic > two-plane construction where the DUT is placed inside. The advantage: > the field is directly related to the input voltage, and an absolute > value can be obtained. > > Ach! Was is das? I've never heard of a TEM cell. What does it stand for? > > --Lawrence > > > ----- Original Message ----- > From: "Jochen Feldhaar" > To: "Lawrence Lile" > Sent: Monday, May 27, 2002 5:31 AM > Subject: Re: [EE]: Sensitivity training for Engineers > > > > Hi Lawrence, > > > > answers below... > > > > Lawrence Lile schrieb: > > > > > > Jochen Schreib: > > > > > > > > > > > you are very correct in guessing that receiver testing is done at VERY > > > > low amplitudes of the test signals. Normally your RF generator is a > very > > > > heavy piece of equipment (both shielding-wise and price-wise) because > it > > > > has to be able to do +20 dBm (100 milliwatt) and also -130 dBm (less > > > > than a picowatt) out of the same assembly. > > > > > > Aha! Any sources of such equipment? My company might actually spend > the > > > money for it. > > > > Some used equipment by HP or Marconi should do the trick, I have here a > > Marconi 2017 that can be modulated in AM, FM and pulse. Most others will > > also have this capability. Buying used will be below 1000$US, but your > > frequency should be included in the range. Some generators start at 100 > > KHz, some at 10 MHz. > > Using a low cost "function generator" for this application will radiate > > too much from the plastic enclosure of such units, unless you get a good > > one e. g. HP 3325, but you will need an external attenuator for a > > function generator. > > > > > > > > > > > > > Also there are different methods for determining the quality of > > > > reception. > > > > > > I'll use a go-nogo method - if my reciever can lock onto the time code > > > signal after transmitting XX number of time codes, say 4 or 5, then it > is > > > good. it would take 10 minutes to transmit 5 time codes. Eventually I > will > > > find a minimum level and set it fixed, so that a production go/nogo test > > > could be performed. Is it sensitive enough, or not? > > > > Definitely the way to go, if the time is available! > > One gotcha: the relative signal lavel must be the same for consecutive > > batches of DUTs... > > > > > > > > > A digital receiver for Time standard signals sometimes has a lot > longer > > > > observation and acquisition periods - not so easy. And also your > > > > "natural" RF signal should be blocked effectively, or your results are > > > > worth less than zilch... > > > > > > > > > > Another reason this has to be done in a well-shielded cage. I am > thinking > > > about a two-layer box, the outer layer being copper screen, the inner > layer > > > being maybe steel, rivetted together and using EMC tape all over the > seams, > > > with an EMC gasket on the door. I have seen such an arrangement at a > > > manufacturer that tests some kind of RF equipment on their factory > floor. > > > > Ja! This is cumbersome, will take some days to build, some cash, but if > > it is well done, you can also perform other tests in there. > > As signal coupling device, I recommend a TEM cell, some kind of rhombic > > two-plane construction where the DUT is placed inside. The advantage: > > the field is directly related to the input voltage, and an absolute > > value can be obtained. > > > > > > > > Mit gern, > > > Lawrence > > > > > > > Hope this helps - more info available if not.... > > > > > > > > Greets > > > > > > > > Jochen Feldhaar DH6FAZ > > > > > > > > Lawrence Lile schrieb: > > > > > > > > > > How do people go about testing sensitivity of radio recievers? I've > got > > > a > > > > > batch of clocks that recieve a time code over WWV, and some of them > > > "seem" > > > > > to be less sensitive than others. Some nights I'll see 5 out of 5 > lock > > > onto > > > > > the WWV signal, other nights only 1 out of 5. I don't have any > > > calibrated, > > > > > predictable way of saying "Yes, this reciever is sensitive enough". > > > Niether > > > > > me nor my supplier has a good way to do this. > > > > > > > > > > I am imagining a transmitter, suitably armored in a really good > faraday > > > cage > > > > > to prevent problems with the FCC, that I can turn way, way down in a > > > > > measurable way, and test whether the reciever modules can pick up > the > > > > > signal. I think this would be quite difficult to build. A > transmitter > > > that > > > > > fakes a WWV signal and transmits 100 mW would be trivial to build, > but > > > how > > > > > about one that only transmits 0.01 microwatt? I could see building > the > > > > > transmitter in a separate can, moving the 100mW signal through a > coax to > > > an > > > > > attenuator, then through another coax into another can holding my > > > product. > > > > > > > > > > What is actually a reasonable level of power that a reciever in the > US > > > might > > > > > get from a transimtter in Colorado? It must be vanishingly small. > > > > > > > > > > Hmmmm..... > > > > > > > > > > -- Lawrence Lile > > > > > Sr. Project Engineer > > > > > Salton inc. Toastmaster Div. > > > > > 573-446-5661 Voice > > > > > 573-446-5676 Fax > > > > > > > > > > -- > > > > > http://www.piclist.com hint: The PICList is archived three different > > > > > ways. See http://www.piclist.com/#archives for details. > > > > > > > > -- > > > > http://www.piclist.com hint: The PICList is archived three different > > > > ways. See http://www.piclist.com/#archives for details. > > > > > > > > > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > > -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics