Author Topic: Rigol DSA815 dead LO for the 900-1500MHz range  (Read 1631 times)

0 Members and 1 Guest are viewing this topic.

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Rigol DSA815 dead LO for the 900-1500MHz range
« on: December 13, 2024, 03:38:21 pm »
hi there,

Has anyone ever reverse-engineered the three LO VCO's in the Rigol DSA815?

A couple of years back my 815 failed for the 900-1500 MHz range. That corresponds to the #3 LO VCO. eevBlog forum member ted572 has nicely documented this as the LO range 3,1857 MHz - 3,7857 MHz.

In recently acquired a real SA, an HP 6.5 GHz unit, which finally allows me to check on the LO signals. Some measurements proved my assumption to be correct: the two lower frequency LO VCO's produce signal as they should, the top-band VCO does not produce any signal.

<RANT>My 815 is suffering from absolutely abysmal soldering quality. And, believe it or not, C245 and C246 (hope I read the silkscreen correctly) were simply gone... The caps managed to hold on for the 24 months warranty period + 1 month and then Poofff, spontaneous failure.

Of course our beloved Dutch importer did not want to address the issue at the time: "sorry, out of warranty, and we only repair at module level. That would be EUR 1000+ please". I will spare the forum my response. Let it suffice that I will never again buy any Rigol stuff.</RANT>

So, any chance of repairing would be based on someone who has put effort in re-creating the schematics for the VCOs. And even more importantly: knows the component values used?

Any help much appreciated!
Wilko

 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #1 on: December 13, 2024, 07:56:39 pm »
I don't have a Rigol DSA815 (never used one) but can you put up a closeup image of the VCO?

I might be able to help a bit if I can see an image of the VCO, especially if you can provide a measuring scale alongside it or at least identify if the SMD caps are 0603 package as this will help me scale it all.
 
The following users thanked this post: egonotto

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #2 on: December 14, 2024, 01:50:07 am »
I can contribute a photo of the 1st LO section of mine but the component values are unknown... Peculiarily, both the mentioned caps and a few others as well had been reworked right from the factory on this specimen. Is the LED of the 1st LO branch in use lighting up?

Edit: Transistors are BFP450 (Marking "AN") and BC817 (Marking "6C").
Edit2: Added a photo of the rear side of the 1st LO section, containing all the power supply and PLL circuitry. The PLL (ADF4169) is controlled by a Spartan FPGA.
« Last Edit: December 14, 2024, 12:03:32 pm by TurboTom »
 
The following users thanked this post: egonotto

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #3 on: December 14, 2024, 10:56:55 am »
Yes, all 3 VCO range LEDs light in sequence for a full sweep to 1.5GHz.

And some pictures with scale.
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #4 on: December 14, 2024, 01:11:19 pm »
Thanks for the images and the scaling info :)

Can you confirm that there is actually a row of ground vias at the end of each resonator as shown in yellow in the image below? I'd expect to see those there. It may be that they are filled vias so difficult to see them. Don't worry if you can't see them under the solder resist.

Also can you read the part number code on the two varactor diodes shown by the green arrows?

I assume the missing caps are the ones circled in red in TurboTom's image.


You will need to use caps that are well behaved up at 3 GHz as the package inductance will begin to affect the capacitance up at 3 GHz.

Is there a tiny part code stamped on the red circled caps in TurboTom's image? ATC parts often have a part code on them.

The VCO looks like a negative resistance design. To aid in reverse engineering it I can explain how I think it works, but this might take several posts. Ultimately you just want the values for those two caps. I suspect that they could be select on test parts and the same goes for the cap just above the BFP450 BJT (circled in blue). This cap helps define the frequency range over which the BFP450 generates negative resistance. The red circled caps help define how the resonator is tapped into by the BFP450 BJT.

They are probably selected to give the most linear tuning response for the VCO and they also help define the oscillator power in the resonator. I don't know if the Rigol analyser is clever enough to self calibrate its VCOs to linearise them using some form of lookup table that helps provide a linear sweep on wide spans (eg across the full range of each VCO). This would help a lot in terms of achieving close to factory performance when you fit new caps here.

I can explain how it all works but maybe this will just bury the real goal which is for someone to tell you what the actual cap values are?



« Last Edit: December 14, 2024, 01:13:36 pm by G0HZU »
 
The following users thanked this post: egonotto

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #5 on: December 14, 2024, 03:12:02 pm »
Many thanks for taking an interest!

There do not appear to be vias in the resonator. See attached picture please (take via the stereo microscope in case anyone wonders).

First the easy part: the varactors are marked BY.

C247 is unmarked.

I assume the missing caps are the ones circled in red in TurboTom's image.  That is correct.

I obviously cannot check the red-encircled caps as they are AWOL in my unit. But I did take a picture of the corresponding section of one of the other VCOs (see attached). Alas no markings.. That abysmal soldering quality you see is Rigol's doing... It looks awful enough to make your assumption about Select on test, so handfitted components, plausible. It also explains why the caps simply fell of the board, the soldering 'quality' being what it is I mean.

 (If I had ever attempted a design like that in the dark ages when I was in computer hardware design I would surely have been taken behind the shed and shot by my manufacturing engineering colleagues :palm: )

As for whether the Rigol does sofware based linearisation I really have no idea.

Wilko

« Last Edit: December 14, 2024, 04:03:36 pm by wkb »
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #6 on: December 14, 2024, 03:58:16 pm »
Thanks. I'm a bit confused by the lack of ground vias at the end of the resonator. Can you do a continuity check from the lowest varactor diode (anode end) to ground? i.e. is that resonator shape grounded somehow?

It may be that there is a clever way to ground it that is better than a conventional via pattern. I think it has to be grounded because the diode anode needs to be dc referenced to ground in some way. So maybe Rigol have a special type of ground connection here?

Can you also check to see if the right hand end of the cap circled in blue in my image is also at ground? This is effectively pin 3 of the BFP450 BJT. I'm trying to work out how the BJT biasing is arranged.
 

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #7 on: December 14, 2024, 04:23:00 pm »
I checked all 3 resonators by measuring resistance to GND All are dead short to GND. How... I have stared through the stereomicroscope for a considerable time but I do not see how it was accomplished though.

The left hand side of C245 (or C226, for comparison) also appears to be on an etch island without being connected to something. Maybe some stealthy via concept in the Rigol PCBs?

Right hand side of the blue cap is connected to GND.
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #8 on: December 14, 2024, 04:49:00 pm »
Thanks again! I'm fairly certain there will be a clever way to ground the bottom end of the resonator. Maybe copper filled vias have been used here and the top surface of them is so smooth that they can't easily be seen under the solder resist.

The circuit doesn't make sense to me if the resonator isn't grounded as per my initial image using the yellow vias.

It looks like the oscillator uses a classic method to generate negative resistance up at >= UHF using a decent BJT. This involves grounding the collector directly and then using negative bias voltages at the base and emitter and then adding the blue circled capacitance from the emitter to ground (which is the collector). This assumes that pin 3 of the BJT is the collector and it is grounded.


I can show you how the BFP450 achieves the negative resistance (at the base pin) across the full VCO range if that helps? I can also show you how the resonator is arranged. Sadly, this won't be able to predict the capacitor values accurately, but it might give some insight.
« Last Edit: December 14, 2024, 04:50:53 pm by G0HZU »
 
The following users thanked this post: egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #9 on: December 14, 2024, 05:50:31 pm »
Here's an initial way to view that oscillator as a negative resistance oscillator. This is just a crude analysis to get thing going. Ideally, the circuit needs to be analysed as a complete PCB with the layout pads etc.

In the typical (simple) lossy resonator circuit shown below L1 and C1 have resistive losses modelled by R2. In order to make this resonator oscillate, the negative resistance of R6 needs to be enough to offset (cancel) the resistance of R2. This is fairly straightforward :)

However, your circuit is a bit more complicated. I've still simplified it a bit compared to the real PCB but this should still be a good way to look at how it works.

The BFP450 BJT provides the negative resistance as the pink resistor R5, but it can't be connected 'strongly' to the resonator or it will limit the power in the resonator and it will give poor phase noise.

The bulk of the resonator is L2 R4 C2 and C4 (unknown value).
C4 provides a tap point in the resonator to provide a looser connection to the BFP450 BJT. C5 (unknown value) also provides some isolation between the resonator and the BFP450.

Ideally, there will still be enough negative resistance provided by the BFP450 to allow the circuit to start up and the energy in the resonator will build and build until equilibrium is reached. When this happens the BFP450 has hit its limit. I'd expect the power in the resonator to be several milliwatts and there could be quite a high RF voltage at the top of the resonator. Much too high an RF voltage for the BFP450 to cope with but it is protected from this voltage by the capacitive taps in the circuit.
 
The following users thanked this post: egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #10 on: December 14, 2024, 06:25:00 pm »
Because the collector is at AC ground, it's possible to use the manufacturer's s-parameter model for the BFP450 and re-arrange the circuit to make it look more familiar as a Colpitts oscillator.

In the image below I've represented the blue circle cap as C3. I've guessed the value as 1.8pF but it might be a bit different to this in reality. This is fitted across the collector and emitter pins of the BFP450 but if the collector is ground then it is more intuitive to redraw the circuit as below.

C7 is the internal base to emitter capacitance of the BFP450. Because this is already there inside the Q2 model, I've given C7 a value very close to zero in the simulation.

You can see that this now resembles a Colpitts oscillator with the capacitor based feedback. If I simulate it, the simulator shows that negative resistance is generated across a huge frequency range extending to beyond 4.5 GHz.

So the BFP450 can do its job as a negative resistance generator :) I've plotted the series resistance looking into the base rather than Rp but you can see the resistance at the base is negative. Don't pay too much attention to the value of the negative resistance as the circuit really needs to be modelled in a way that includes the PCB layout.
 
The following users thanked this post: egonotto

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #11 on: December 14, 2024, 08:04:07 pm »
The varactor diodes are most likely Infineon BBY58-02V, assuming the package is an SC79 and the marking is "8". The PIN diodes are BAR63-03W (SOD323, marking white "G"). The SOD323F diode with the "US" marking probably is a BZX84J-C6V8 zener. Unfortunately, none of the capacitors is marked.

Edit: I found a slightly better photo of the 1st LO section of the DSA815 in my archives, this one even just barely shows the blind vias at the "T" section of the resonator striplines, confirming they are hard-linked to the underlying ground plane.
« Last Edit: December 14, 2024, 08:26:16 pm by TurboTom »
 
The following users thanked this post: egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #12 on: December 14, 2024, 09:08:03 pm »
Thanks for the latest image that shows the ground vias. They appeared to be missing in the other images but they must be there.

In the marked up image below, the C247 cap at the top will be chosen to give negative resistance over the required frequency range. The value might get adjusted during board testing by Rigol to adjust the output power of the oscillator.

The (missing) C246 will also have some impact on the output power of the oscillator and it will also have a small impact on the tuning gain of the VCO in terms of MHz/V.

The other missing cap C245 will probably have the biggest impact on the tuning gain of the VCO in MHz/V. It may need to be selected on test to offset any tolerance issues with the varicap diodes and also with the etching tolerance of the resonators. Adjusting this cap value may mean that the other two caps also need to be adjusted to help keep the output power and tuning range close to the design goals.

I suspect that the analyser can measure the tuning voltage (vs frequency) of each VCO during a power on self test routine.
 
The following users thanked this post: egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #13 on: December 14, 2024, 09:14:15 pm »
If you can find out for certain what the varactor diodes are, I can try and model the whole oscillator layout using Genesys and Sonnet EM and maybe predict the approximate values of the missing caps.

It would be nice to know what the typical tuning voltage range is at the test point V1 across the full tuning range of the VCO. I'm assuming it will be something like 2V to at least 10V, maybe much higher. Could this be documented somewhere in a service manual?
 
The following users thanked this post: egonotto

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #14 on: December 14, 2024, 11:59:57 pm »
The varactor diodes are most likely Infineon BBY58-02V, assuming the package is an SC79 and the marking is "8". The PIN diodes are BAR63-03W (SOD323, marking white "G"). The SOD323F diode with the "US" marking probably is a BZX84J-C6V8 zener. Unfortunately, none of the capacitors is marked.

Edit: I found a slightly better photo of the 1st LO section of the DSA815 in my archives, this one even just barely shows the blind vias at the "T" section of the resonator striplines, confirming they are hard-linked to the underlying ground plane.


My varactor diodes are marked BY, not 8. Whether that means anything, or just a different vendor was used, who knows.

Soldering on your PCB is a helluvalot better than on mine by the way.
 

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #15 on: December 15, 2024, 02:35:30 am »
If the marking of the varactors is "B", which can easily be misinterpreted as "8", considering the size of the packages, the component may be a BB555-02V. Yet, this diode requires a higher control voltage of up to 28V. On the other hand, its minimum capacitance is way lower than that of my previously assumend component, the BBY58-02V (2.1 vs. 4.7pF). The BB555-02V would also better fit the frequency range of this oscillator. It's highly likely that Rigol used an Infineon component here since the other diodes and the transistors are of this origin as well.

G0HZU's hint that these capacitors are actually used to adjust the oscillators during initial testing at the manufacturer makes good sense: On my unit, for the highest frequency, all three capacitors (series, parallel and feedback divider) are hand-soldered while the mid range oscillator only has got the feed back divider cap "patched" and the low-range none at all. The higher the frequency, the more critical parasitics and component tolerances get, so voila...

It's strange that your DSA815 main board is of such a bad soldering quality, actually all of the Rigol gear that I had a "peep" into, was of very decent build quality, I never found any reason to complain about that. Their firmware is a different story, though ;). Yet, that's not the point here... You may have received a lemon that slipped through the quality control, but obviously, that won't console you.
« Last Edit: December 15, 2024, 11:14:56 am by TurboTom »
 
The following users thanked this post: egonotto

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #16 on: December 15, 2024, 10:38:02 am »
Better visible under the stereo microscope but to me the varactors are marked BY. Really a B, not an 8

As for a 'lemon' Rigol, I agree. But buying from a company that has no adequate quality control is just crap shoot. My conclusion is simple "Rigol? Nie wieder!"
 
The following users thanked this post: TurboTom

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #17 on: December 15, 2024, 11:12:39 am »
Better visible under the stereo microscope but to me the varactors are marked BY. Really a B, not an 8

As for a 'lemon' Rigol, I agree. But buying from a company that has no adequate quality control is just crap shoot. My conclusion is simple "Rigol? Nie wieder!"

You are right, this is clearly a "B". I had to use the old photos that I took of my own unit to identify the components since the instrument now is  located at our "production site", some 400km from me, so I cannot verify with the real thing. I've got many, quite good photos, also of the other, similarly configured oscillators distributed across the PCB, and I was drawn between "B" an "8", and then I first found the (datasheet of a) varactor with the "8" marking, and I thought "that's it"... But your microscope shot clarifies the situation.

I very well understand your feelings towards Rigol. Fortunately, there are enough other sources for such instruments that may suit you better. Anyway, receiving an instrument paid for out of the "hobby wallet" or at least with personal money, that turns into a dud like that, hurts and affects future purchase decisions. I've also been through that and try to avoid certain brands (unless the deal is too good to let pass...). I hope the repair attempt will turn out successful!
« Last Edit: December 15, 2024, 11:31:07 am by TurboTom »
 

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #18 on: December 15, 2024, 01:02:17 pm »
I think there's still a lot to be learned from studying the design challenges associated with this type of VCO.

The printed resonator probably looks like 0.8nH at 3.5 GHz. Each diode will probably have 0.5nH package inductance and the C245 cap and its associated PCB grounding might add another 0.4nH.

So that means about 2.2nH inductance in the resonator. This only requires about 1pF capacitance for resonance at 3.5 GHz. Therefore, some more study is required as below...

A big clue to the challenges involved for Rigol is that they had to use tiny 0402 parts for C245 and C246. I think this is because C245 acts as the dominant cap in series with the varactor diodes. So it probably has to be >2pF.

There only has to be 0.5nH inductance in series with 2pF to make the 2pF cap look more like 4pF at 3.7 GHz. So tiny parts are needed here or they can misbehave up at 3.7 GHz... :)

Something like the Kemet HiQ-CBR series in 0402 package size will be needed. There are 0402 caps available from ATC and PPI that would also be fine here but they will cost more.

The higher the value of C245, the higher the VCO tuning gain of the VCO in MHz/V. It's important to try and achieve roughly the same VCO tuning gain (Kvco) as the original design because this affects the loop bandwidth of the PLL used for the first LO.

That's why it would be nice to know the correct tuning voltage at 3.186 GHz and at 3.786 GHz. If this is known, then it helps a lot. However, C245 needs to have similar package inductance to the original Rigol part. This needs to be really low and this is why there isn't much point trying 0603 or 0805 sized parts in this circuit.





 
The following users thanked this post: wkb, egonotto

Offline wkbTopic starter

  • Frequent Contributor
  • **
  • Posts: 999
  • Country: nl
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #19 on: December 15, 2024, 02:16:48 pm »
Thanks for your insights with regards to the physical size of the capacitors involved. I did not realise that stray inductance is that important, on the other hand I should have known that already  :-//

So, I went and soldered a strand of wire onto the V1 varactor control voltage line. The outcome of what I saw on the o'scope is in the attached screen captures.

I captured both a full 9kHz - 1.5GHz sweep as well as the sweeps per VCO range. I guess the filenames of the attached captures are self-explanatory.

That all said: I admit, I am now quite confused  :-/O
 

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #20 on: December 15, 2024, 07:25:53 pm »
Wow, that's quite interesting! So the varactors can be driven up to ~32V maximum, which is beyond their specified reverse voltage of 30V but below the "maximum reverse voltage" with >5kOhms series resistance, which apparently is lower in the oscillator circuits (10 Ohms + 26.1 Ohms + whatever the source resistance of the charge pump interface circuit is).

Another peculiar detail is that the PLL IC that Rigol used in the DSA815 (ADF4169) is capable of generating "waveforms" in the frequency domain by itself, i.e. it can provide the frequency sweep without additional activity of the FPGA. That would easily explain the fast control voltage rise when the inoperational third oscillator branch gets selected and the PLL senses a "too low" frequency.

At the maximum varactor drive voltage, the capacity of the BB555 is round about 2pF, so the anti-series arrangement of two of these diodes result in ~1pF as a minimum. Parasitic inductance of the SC79 package is specified as 0.6nH. This matches G0HZU's figures quite accurately.  :-+

Assuming that the three VCOs are operated over approximately the same control voltage range, we should expect someting like 7V @ 3.186 GHz and 20V @ 3.786 GHz.
« Last Edit: December 15, 2024, 07:34:29 pm by TurboTom »
 
The following users thanked this post: egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #21 on: December 15, 2024, 08:45:15 pm »
Yes, it's definitely getting more interesting now...

It's probably worth mentioning how this circuit will work in the large signal condition. Up until now, I've been looking at small signal 'start up' conditions.

Once the VCO starts up, the energy in the resonator will build and build and there will come a point where the increasing drive signal to the BFP450 base will cause the BFP450 to conduct on less and less of the RF cycle because it will be reverse biased for most of the RF cycle once the amplitude of the RF cycle becomes large. This means that the tradeoff between increasing drive level leading to reduced conduction angle will mean a form of equilibrium will be reached in the large signal condition.

You can think of the BJT as 'tapping' a brief burst of energy into the resonator once a cycle and this maintains equilibrium in the large signal condition. A bit like briefly tapping a bouncing ball to keep it bouncing up and down at the same amplitude every bounce.

I'm not sure what this means for the value of the series unknown cap C246. The value of this cap will influence the output power level of the oscillator, but if it is too big it will load the resonator during the crucial startup (small signal) phase.

It might be worth trying C245 as 2.2pF and then experiment with values from 0.8pF through to 1.5pF for C246. There is effectively a small amount of additional capacitance across the shunt C245 cap in the form of the square PCB pad that connects C245, C246 and one of the diodes and L218. So C245 may end up being a bit less than 2.2pF. I'm really just guessing until I model it all better.

The bias chokes L218 and L211 will presumably be chosen to be close to self resonance up at about 3 GHz so they should appear fairly invisible to the rest of the circuit. They might have a value of about 15nH for example.
 
The following users thanked this post: egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #22 on: December 15, 2024, 08:56:03 pm »
If C245 is chosen to be too large it will begin to isolate the BFP450 too much as it is part of a capacitive divider in the resonator. There is a double whammy effect here because of the package inductance of C245. This will make it look even larger up at 3.5 GHz. So none of this is a trivial task. It may well end up being a case of trial and error.

C246 needs to be chosen to give reliable startup (and correct output power?) and C245 needs to be chosen to achieve the target VCO tuning gain in terms of MHz/V. There will be some interaction between both caps and it looks like Rigol had to use select on test caps here. So don't expect to get it right first time even if someone knows the typical values for these two caps :)
 
The following users thanked this post: egonotto

Offline TurboTom

  • Super Contributor
  • ***
  • Posts: 1471
  • Country: de
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #23 on: December 16, 2024, 12:37:03 am »
Here's the complete schematic of the highest frequency oscillator branch including the bias circuitry. Interestingly, Rigol uses the LED to bias the two current sinks for the base and emitter currents of the RF transistor. In order to disable the oscillator branch, this LED is simply shorted out to the negative supply rail by Q204. The zener D206 provides the level shifting so it can be operated from digital circuitry supplied by a positive rail. Via the emitter current sink (Q206) and R247, the corresponding PIN diode gets positively biased, so as soon as the selected oscillator branch gets enabled, the corresponding RF switch also conducts the signal. Neat and simple design!

P.S. Vee is probably in the ballpark of -5V, considering a 6.8V zener is used for level shifting, and assuming a 3.3V digital rail to power the FPGA I/O.
« Last Edit: December 16, 2024, 12:42:15 am by TurboTom »
 
The following users thanked this post: wkb, G0HZU, egonotto

Offline G0HZU

  • Super Contributor
  • ***
  • Posts: 3238
  • Country: gb
Re: Rigol DSA815 dead LO for the 900-1500MHz range
« Reply #24 on: December 16, 2024, 05:35:29 pm »
That's a really useful diagram thanks!

I've added some waveform info to it as below. The green circle highlights the main resonator. At startup, some energy from this resonator goes into the BFP450 as a forward wave and because the BFP450 is a negative resistance generator, the reflected wave back out of the BFP450 will be in phase and slightly bigger than the forward wave.
After a very short time this energy gets filtered by the resonator into a sine wave at its resonant frequency and because more energy returns from the BFP450 than goes into it, the resonator starts building up more and more RF energy at the resonant frequency. Eventually, the BFP450 can't generate a 'bigger' reflected wave than the incident wave because it has hit its limits and this is when the resonator will have several milliwatts of power at the resonator.

The more power at the resonator, the better the phase noise and the more selective the resonator is (i.e. the higher its loaded Q) the better the phase noise will be.




 
The following users thanked this post: wkb


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf