1GHz VCO using Transmission Line Resonator

[00:00:06]All right, deja vu. Um I'm going back to my Agilent uh RF generator, my 4 GHz Agilent.

[00:00:15]Um and um what I want to do is to start looking at these circuits because I was able to get schematics for the device.

[00:00:26]And so now we can actually see what's going on.

[00:00:30]So, today I want to talk about the uh reference board, which is the problem board that I had in my machine. This is a card that I bought as a replacement and found out that it's not a replacement, it's a different model.

[00:00:43]And it's not compatible. [clears throat] Uh even though they have the same pinout and everything, you plug it in and it says, "Nope, I don't understand that card. Can't use it."

[00:00:51]Um so, we're going to be looking at the 1 GHz VCO on this board.

[00:00:57]And it is right here.

[00:00:59]Uh it it uh goes under the metal can right here.

[00:01:04]And uh it is uh easy to understand this circuit. So, if we can understand this circuit, then when we go to the other revision, um it's a little more complicated, but a lot of the ideas travel over.

[00:01:24]So, if we can understand this one, we will be a leg up. So, uh this is just a simple voltage-controlled oscillator.

[00:01:32]Uh it does its voltage controlled using varactor diodes to change um they're basically a voltage-controlled capacitor.

[00:01:41]Um and so, this thing needs to be exactly 1 GHz.

[00:01:46]And so, you need to get it close, and then there's a feedback loop that will grab that VCO line, grab the control voltage, and it will go this phase lock loop, so it will become exactly 1 GHz. The way that it does that is it has a 10 MHz reference, which is very, very accurate.

[00:02:09]So, it takes the output of this 1 GHz oscillator.

[00:02:15]It runs it into a gate array, which is this right here, divides it by 100. So, it takes the 1 GHz, divides it by 100, it gets 10 MHz. So, we have 10 MHz to 10 MHz, it goes into a phase lock loop, and then we can phase lock this to exactly uh the the phase relationship to this 10 um MHz over here, and we have a very accurate 1 GHz.

[00:02:41]All right, so we need to get close, though. So, you might say, "Well, we'll put in a 1 GHz crystal." Well, they don't make 1 GHz crystals.

[00:02:50]Um and so, we're going to figure out a different way to do that. And so, the way this one is is is this little pipe right here. It's a piece of coax. It's a rigid coax.

[00:02:59]It's a um I forget the RG number, but it's a.141 in polypropylene filled um coax. It is 96 mm long.

[00:03:15]And if you take 96 mm, and you take the dielectric constant of maybe say.69, uh.68,.69 for the the dielectric inside the coax, you figure out that this thing is very close to um 1 GHz half wave. So, this is a half wave, okay?

[00:03:42]Um it's shorted on this end.

[00:03:45]So, it's a shorted half wave, okay? So, a half wave is up and back, okay? So, if we have a short here, then we will also have electrically a short here.

[00:03:58]And then in between we can resonate. So this acts as a resonance a resonant cavity with an input a very low input impedance and that will be that will be key. We we need a very low input impedance to our to our resonator. So just think of this as like an organ pipe pipe organ and this is just a pipe cut to exactly the right frequency and the waves will rattle back and forth in there at exactly 1 GHz. So that's what that's the job of this.

[00:04:26]Now it's going to be a little bit high and we will learn on our other uh machine our other board that they were targeting 1.05 GHz. That's where their starting point was and then they used the VCO to pull that down to 1 GHz. They start a little high and they pull it low. This one also is cut to be about 1.09 GHz I think. Um if you do the math. Anyway, that's my best guess. Anyway, basically just think of it though as a tube that resonates at 1 GHz, okay? All right, so let's take a look at the schematic and you're going to go oh god.

[00:05:08]Uh it's too complicated. Um so we have one transistor that's our oscillator and we have this funny little drawing over here on the side which is our resonant uh coax.

[00:05:24]All right, so this is our little piece of coax. It's labeled here. It looks like 10 cm, but I think that means I think they they misread it when they when they typed it in or uh but it's supposed to be 100 mm.

[00:05:39]And then it's cut to length, right? So it ends up being about 96 mm.

[00:05:43]Anyway, uh that's the resonant cavity.

[00:05:46]Uh the resonant cavity is tuned with capacitors.

[00:05:51]So, there's 1 2 3 4 5 capacitors and we'll look at those.

[00:05:56]Um, but uh, they are not all loaded. Some are loaded and some aren't loaded depending on the um, frequency matching. So, these are This is cut to length and these are tuned.

[00:06:12]So, the whole thing works at exactly a gigahertz and then you need to go plus or minus from there, which is the varactor diodes. So, the varactor diodes are up here, okay? So, we have varactor diodes. So, that's basically the resonant cavity. It's the half wavelength capacitors tuned with the varactors, with variable variable capacitances. And then we have this thing. So, let's talk about this thing.

[00:06:34]It may look a bit strange.

[00:06:38]I did a video once. I'll try to link it down below on common base amplifiers.

[00:06:43]A common base base amplifier looks like this.

[00:06:47]You have a uh, NPN transistor that is uh, has the base grounded.

[00:06:53]And we have negative voltages here and positive voltages here.

[00:06:58]So, you can imagine the plus comes through here, it goes through the arrow in the right direction and then comes out here. So, it is biased on, right?

[00:07:07]This this uh, uh, these two voltages bias this transistor on at a certain point, okay?

[00:07:15]And um, this is the input and this is the output.

[00:07:23]And you're going to say, "Hey, Mr. Wizard, that's not right. We can't go through this transistor. It's in backwards."

[00:07:30]Well, um, yes, you can, okay? As long as the transistor is being conductive, it's on, then a little small signal, a little small AC signal can go through. So, it's DC biased on, but AC biased it can actually go through in this direction.

[00:07:49]All right.

[00:07:50]So, again, that's in the other video if you want to go watch it. But anyway, a little bit of wiggle here is a big wiggle out here. It amplifies, okay?

[00:07:58]And uh so, this is what we have, okay?

[00:08:02]The input, we basically have a tuned circuit, like an LC or a crystal or a resonant cavity or something. We have this tuned resonant circuit over here.

[00:08:15]So, it's going to wah-wah-wacka-wacka on a particular frequency, all right?

[00:08:20]And um yeah.

[00:08:23]So, let's go back.

[00:08:26]And here it is.

[00:08:28]Get it closer here. So, uh we biased it on by a plus 14 V. It goes through some bypassing.

[00:08:38]Here's some bypassing. Here's some series resistance to set the current level.

[00:08:43]And then it goes into the transistor and then it comes up this way. It goes through some resistors and stuff to a negative voltage, minus 14. So, plus or minus 14 bias this transistor on.

[00:08:54]There's some ferrite beads and there's some inductors and there's some other things in here that um uh make this happy, all right?

[00:09:04]It wants to make sure that we bias this thing DC.

[00:09:08]So, this inductor here and this inductor here allow the DC biasing, but as the 1 GHz goes flying through here, the 1 GHz can't go up. It it it It's blocked going in the up direction, okay?

[00:09:21]And so, that's what those two things do.

[00:09:23]And then it comes out and it's going to be pretty small.

[00:09:26]And so, it gets um amplified a second time, okay? So, this this amplifies it just enough to oscillate.

[00:09:34]And then this uh amplifies it to be used elsewhere, all So, in our device here, let me zoom down.

[00:09:51]All right. So, again, here's our resonant tube. Uh, here's a tiny little SOT23.

[00:10:01]And that is a dual diode package, and it's actually a dual varactor package.

[00:10:06]Guess the two varactor diodes are actually in that little package there.

[00:10:11]And, uh, that resonates and it goes zoom-a-zooma.

[00:10:15]It gets amplified by the transistor here. Uh, you can see like the input the string of resistors coming in, and then there's a string of resistors going out.

[00:10:24]So, these are these are the two paths to bias this thing DC-wise. Once you get this thing going, we have this other little transistor down here, or it might be a mimic or something transistor. Uh, yeah, it's a mimic cuz it's labeled U instead of Q.

[00:10:39]So, this is a little mimic amplifier that sends it everyone else on their merry way.

[00:10:45]So, now we know how this one works. We have a better understanding how the other one works. So, we're going to do the other one in a separate video.

[00:10:55]Uh, before I leave, I need to talk about, um, this particular thing. You're going to say, "Well, there's no feedback. There's no positive feedback to make it oscillate." Okay? There's nothing from the out to the in, and there's no way to get positive feedback in this thing.

[00:11:11]And, uh, that's kind of some of the magic of RF design is a lot of times the stray capacitance or the Miller capacitance, um, the PC board capacitance, there's extra capacitances that aren't shown on a simple diagram like this.

[00:11:27]And I believe it's the Miller capacitor that allows this thing to oscillate, uh, but that's, uh, out of the scope of this video.

[00:11:34]Uh, But you might want to investigate that on your own. A lot of times you will see oscillator circuits and you say, "Oh, the guy drew it wrong. He forgot the feedback path. He forgot blah blah blah blah blah." No, he didn't forget it. It's just not necessary that the inherent capacitance of the transistor actually makes it go. And think about this, at 1 GHz, it doesn't take a lot of capacitance. So, just a tiny tiny tiny little bit of capacitance will make this thing go.

[00:12:07]>> [music]