Well, good morning. It's Christmas Eve here in Seattle and it's just below freezing so maybe we might be a little bit snow this evening. White Christmas tomorrow. Anyway, I Wanted to do a quick video to submit to Dave Jones a Eevblog, hopefully for inclusion in his holiday videos.

Now the video in question I'm doing here is responding to a viewers question on my my channel where they looked at a power meter I had and wanted to know with that power meter how they would go about working out the sensitivity of a crystal detector. So here is a set of crystal detectors that I have. This one is a 9470 beam and it's an N-type connector and it goes from ten Megahertz tating gigahertz. The 8472 B is here in the middle and it's a SMA type connector and it goes from ten megahertz to adding units as well.

and then on the right here is an 84 73 C and it is a 3.5 millimeter connector which is physically the same as an SMA but they're actually different because of the tolerances. So if you're ever using these things, mostly people have SMA cables so it's a good idea to have a little saver in the connector there so that you wear out this saver and you could throw that away and replace it anyway. this is a 10 megahertz to 26 and a half gig crystal detector. Now what this crystal detectors do is they make use of the square law area of these Schottky diodes and what that means is that below a certain power when you put the power into a crystal detector, the slope of a line that shows the action or shows the correlation between power and the output voltage becomes proportional and linear and the viewer let me move these out of the road here and we'll zoom out a little bit.

The viewer wanted to know how to workout. Let's just let me move that. How to work out what K is here and K is the sensitivity of the the detector and so if we look over here, you can see how the power input relates to the detector law and what we're trying to do is get a linear region here and you can see this linear region part here. and so they call this based off this equation.

Here they call it square world region because Alpha here becomes 2 and hence squared. So when you work out what the equation is, it's the voltage out of the crystal detector is the sensitivity Chi square root of the power squared which is just the power. So you become linearly related to the power and you can see the voltage. You can see that the lines up here are very linear based on the power and the frequency through Giggler tube for the voltage that's being detected and so he wanted to know how to specifically get that calculated voltage for Chi.

Now if we jump out here and take a look at the specs, you'll see that typically the sensitivity. So let me move that down. Typically sensitivities accorded in millivolts per micro watt, and so the K value as a unit of measure as far as I'm aware, is not well used. So anyway, on most of the data sheets, you'll see sensitivity as milli volts per micro watt in that square law region.

and what we want to do is to take these detectors and go ahead and calculate their sensitivity. So to do that, I'm going to need to use two pieces of three pieces of gear actually, so let's zoom out a bit so you can see that sent these things in a row. The three pieces of gear We're going to be a signal generator that's capable of delivering minus 20 DBM at and we'll pick a middle frequency of 10 gigahertz. I'm going to use a pal made on that enables me to check the power at the measuring point and this will be the measuring point here where I have my cable connected to the N-type connector here and I want to set this to be minus 20 DBM and then we'll use a digital multimeter where we can connect it to the BNC connector here on the end, which will show us the voltage that will occur.

So let's go set this up. The first thing we're going to go do is set up our set up the signal generator. Okay, so the signal generator I'm using is my 86 73, B and I primarily use this as a local oscillator for a microwave converter, but I'm going to use this at this particular case because this has a Vernier control on the output and it enables me to you know very finely adjust the power. I Thought about using my 80 340b which I have here, but unfortunately at Minus 20 DBM is an attenuator step so it's very difficult to get the accuracy between the two.

Get minus 20 DBM. So let's go in and we'll select the range of minus 10 here. and if you see the Vernier if I let me put the output on and now you see the the vernier here. you can see in the little dial that as I move that back and forwards we're going between minus 10 from that range so it's minus 20 DBM up to 3 DBM So minus 10 DBM would be there and that lets me go and get the accuracy at that point I could also go down a range and then just sit here and play around here Now I Want to set my output to be 10 gigahertz? We have our 10 gigahertz there.

We have our automatic level control is being internal so that's a good enough start for us to go and start measuring the measuring the power that we're going to deliver out of that end type connector. The next thing we need to do is set up our power meter. So here is our power meter. So the first thing we'll do is we'll set will do a pre select so we'll make sure that it's all set up.

We're gonna go in and configure it. So these power meters in this bigger case, it's a Need 44 18 B You can use the 8400 series power sensor and this is an 84 81 a power sensor that goes from a hundred megahertz to or gose actually from like 10 megahertz up to 18 gigahertz. And you can see that I have the power factors here. Now this has been off to case ID and it's been calibrated so it should be reasonably accurate given that these things are an open-loop power meter.

So unlike the four three Twos or the crystal detectors which give you an absolute value of the power, they're actually measuring these work on the accuracy of the underlying power reference here. So this is a 50 megahertz 0 DBM pallet of 1 milli watt power reference. So we're gonna put this on here and then we're going to zero that so that this thing knows what the 0 DBM is. So we'll just screw this long here now.

I'm gonna go in and select a table and we go and make sure that you can see here that I have my 84 81 a table. So that's taken all those calibration figures off the side of the unit I've entered those in. So now that we've got that set I can come over here and go to 0 and we're just zeroing out the thermocouples. We've got one that will measure the ambient temperature.

We've got one that will measure the RF that's coming in. and so now we can just go in and calibrate that you can see that the power reference turns on here and there we go. we're calibrated. So now the next thing I need to do is actually put a frequency in and I know that we're gonna be working at Tango.

Good! So let's go in and set that down to 10 Hertz And here you'll see now the calibration factor is set for 94.5% and this is how the unit knows what the losses are going to be in the detector. So now that we set up there, what I need to do is just go and hook this up to the to our cable and will now measure and get our minus 20 DBM So you can see now that I'm measuring -12 So I'm just gonna grab that Vernier and start taking that down until we get to about minus 20 and we're gonna try and get as close as we can. There we go. That's nice and it'll bounce around a little bit there, but you know, a couple of hundredths of a DBM I'll take that.

Alright, so now we just need to go back to our crystal detectors and we can start doing our first measurement. Okay, so we had minus 20 DBM coming out of here. so the next thing for me to do is to hook this up to the crystal detector and to the DMM and that's a Rygel damn 30, 15, 30, 58 So let's just hook this up and now we're going to be reading a voltage here somewhere in the middle volts. So what I have is I Have a little app here that I wrote to pull the data off the off the Rygel because it's in a sort of an inconvenient place.

And so here we have Six Point Six Nine, Six Point Seven million volts. So let's just say that that's going to be our value Now if you remember the sensitivity was going to be or the voltage spec was in millivolts per microwatts and so it's going to be six Point Seven million ten microwatts. So because we're at minus 20 DBM so we got to get point Six seven and if we come back here and zoom in here, you'll see that for sensitivity, it's greater than 0.5 millivolts per microwatts. So you know we're greater than that at Point Six Seven.

So we're in spec and that's given us the sensitivity. Now if we want that K factor, we should go in and bring all those up to volts and and what. So it's a thousand and a million. So basically we're just going to multiply this number when you do the map by thousand So let's go and see and it's about six hundred and seventy.

So to work out from this formula here, using the crystal detectors in that sensitivity region to get Ty was fairly straightforward and for this particular sensor that I have here, it's about six hundred and seventy. Now, these are not metrology grade connectors. There's some loss in the cables, all that sort of thing, but that takes you through the process of how to go and get that. So I could go and do it again using each of these or I could go and try different frequency.

Remember this sensitivity will change based on the frequency. not by a lot, but it will change. So if you're all going and needing this value for 18 gig, you need to do it. At 18, you can measure it and then you can work back anyway.

I Hope you found this interesting and I hope this video makes it onto the Eevblog channel. If it doesn't, it'll still make it onto my site. As Dave said anyway, check below where my channel is TG soapbox and come on over! I have a bunch of videos about old Hewlett Packard test gear and or old Teske and how you can use it and how to repair it anyway. Look forward to speaking seeing you in the comments catch you later bye.