[SOLVED] voltage regulation in RF circuits (VCO)

[msis]

I am trying to connect a digital mixer (https://www.analog.com/en/design-ce...ion-boards-kits/EVAL-HMC798A.html#eb-overview) and a digital multiplyer (https://www.analog.com/media/en/technical-documentation/data-sheets/hmc578.pdf) to a voltage controlled oscillator (VCO) ...

Am I write to assume that this shouldn't be possible via coaxial-connections?

Is it possible via waveguide connection?

would it also be possible to connect a VCO operating on 3.3V to the 2 boards above which are operating on 5V?

thx in advance

 

[Georgy.Moshkin]

There are some commercially available coaxial cables working around 30 GHz, i just found some cables Hubersuhner and Astrolab, try google for it 30 ghz sma or something like that. I am sure some cables can be used at higher frequency with some mismatch/loss of 3-6db. Also there are some tiny sma-sma adapter connectors without cable.With waveguide you still need a transition (sma to waveguide). Also it can be done through custom board with GCPW line and SMA connectors soldered.

If you are using power adapter (220v), it is better to use only one, for example 12v adapter, and then put two regulators for 5 and 3.3v. For expensive stuff I would use power bank (battery) + linear regulators, especially if poking around with some measurement devices which are powered from 220V AC.

 

[msis]

I have such cables and it doesn't work ... as soon as I apply the signal from the VCO (both 5V and 3.3V) my power supply tells me the devices (mixer and multiplyer) don't draw 5V anymore but more current than supposed to

 

[Georgy.Moshkin]

Check all component datasheets, maybe some outputs have DC voltage, or measure each block RF in/outs with voltmeter.
HMC578LC3B in and out is AC coupled, and HMC798ALC4 RF is connected to gorund, so problem must be somewhere else.
Where do you apply 3.3v? Both mmics are single supply 5v.
Also you may power each block separately and check if there is no short circuit.

 

[KlausST]

Hi,

It's not clear which voltage you are talking about. Draw a sketch.

If you are talking about the VCO_control_voltage (to control frequency) then this is rather low frequency and does not need HF wiring. In case you expect HF introducing this signal then a simple RC low pass filter close to the VCO should be sufficient.

Klaus

 

[vfone]

Design a common PCB for all three components (using the same PCB material as is recommended by Analog Devices for their ICs), use microstrip or CPW for RF signals between circuits, and everything should work fine, with less headache (as you get when use coaxial cables).

 

[msis]

trying to answer all in one post

@vfone I am not capable of designing my own PCB with these components. That's why I posted this in the "beginner's section" (just got moved). I had a feeling, it's a simple voltage level question

@Georgy.Moshkin You seem to suggest that the AC coupling, and the grounding respectively, of RFin should make it possible to connect via coaxial connections ... is this correct?

@KlausST I am attaching all datasheets of the 4 devices (all evaluation boards)

I have 2 frequency synthesizers (one operating on 3.3V [attach synthesizer 1], one on 5V [attach synthesizer 2]), a mixer, and a multiplyer

It is not possible to connect either of the synthesizers to the mixer, nor to the multiplyer via coaxial connections .... as soon as I initialise the synthesizers, said voltage dropping and drawing of too much current occurs (as indicated by the power supply)

should it work with waveguide connection? that is the actual question here

 

[Georgy.Moshkin]

should it work with waveguide connection? that is the actual question here

depends on sma to waveguide transition, in some designs it may be shorted.


Potential problems:
1) from synthesizer 2.pdf

The Tx and LO outputs from the ADF5901 contain dc bias voltage and are available on the J8 (TXOUT1), J9 (TXOUT2) and J4 (LOOUT) output SMAs.

from adf5901 datasheet it is unclear, it is only mentioned in evaluation board pdf.

2) synthesizer 1.pdf
schematic shows DC block capacitors. Do you have original evaluation boards or custom boards?

 

[msis]

I have got the original evaluation boards .... I also trie dto use a DC-block on J8 (TXOUT1), J9 (TXOUT2) and J4 (LOOUT) with the ADF5901 board ..

 

[msis]

Ok, I have now managed to connect the synthesizer 2 AUX out to the Mixer LO in

Next I try to switch on the multiplier (connected to synthesizer 1 and mixer)

I get the same power problem described above

Can I not use the Mixer RF as an INPUT ?
can I not connect the multiplier RFout to the Mixer RF as input?

 

[vfone]

The synthesizer and the multiplier have DC blocking capacitors (internal or external), when the mixer have only the LO port AC coupled. The RF and IF ports are DC coupled and may need DC blocking capacitors.
I recomend to test each module separate, for functionality, and then connect all together.

 

[msis]

well I can only test the multiplier together with the mixer bc my spectrum analyser is only good for up to 9 GHz ... I need the mixer for a frequency-downconverting receiver structure before the analyser

Is it that you can only connect either AC coupled outputs/inputs or grounded outputs/inputs, the latter also to be connected to antennas?

Is this my mistake?

 

[msis]

OK, I had not applied any signal from the synthesizer .... I had used the wrong cable

I now believe the AC coupling, and RF grounding respectively, are faulty

it ll take me a few days to voltmeter around to actually check that

meanwhile I will buy that (up to 4.4 synthesizer from them) If that one is faulty, I will know within an hour.

If they actually do send me faulty devices on purpose, I will sue them for £ 5000 next Friday.

I just one such a court case, and at the moment I am winning another ....

 

[msis]

The only question now is do I sue them for £5000 or £10000 .... The latter is still possible without solicitor online here in the UK. And AD has an UK office. The filing takes you about 15 minutes. As I said, I am quite experienced with that. (there is other devices too that they ruined before sending me)

I will send them one last very polite professional email ... that is scheduled for Wednesday, could get delayed a bit, I'd say at the latest Friday morning they have it.

By Monday morning 10th of Aug 2020 at the latest, they are sued. Probably for £ 10000.

All ready made synthesizer plus amplifier in that frequency rage costs about £ 13.000 (I have a quote for that already)

So, AD either sends me their devices or they will give me the money for an alternative solution. Eventually, they would probably be ordered by the court to still send me their devices as advertised (but that may happen only in February, March next year. The £10000 (or 5000) they will have to pay beginning of October.

--- Updated Aug 3, 2020 ---

I tried to communicate with them informally ... I hope they tell their bosses NOW, or at least the person who has authority to decide over £10.000 (strongly leaning toward 10.000)

because my official email will be a bit late. And, if they wait for that they will have to pay something even if they send me all 5 devices anew.

thanks

 

[msis]

wow, now they totally self-incriminate themselves. some senior engineer sends me an email ... is like obvious, he knows they sent it faulty deliberately and is now trying to mislead me ...

I have only tested doubler and mixer ... but, I will test all five devices first .. bc probably they are all faulty .. also I have some other deadline to take care of .... so me sueing them will be a bit delayed .. .I reckon in 1 1/2 weeks time they are sued for £ 10.000 ...

 

[FvM]

If I was equipped which such a case as consultant, I would ask if the vendor warnings about evaluation kits usage have been observed. There are also detailed warranty conditions, by the way.

Example from LMX2595 EVM Instructions, one of the units used in your setup:

Evaluation Kits are intended solely for use by technically qualified, professional electronics experts who are familiar with the dangers and application risks associated with handling electrical mechanical components, systems, and subsystems. User shall operate the Evaluation Kit within TI’s recommended guidelines and any applicable legal or environmental requirements as well as reasonable and customary safeguards. Failure to set up and/or operate the Evaluation Kit within TI’s recommended guidelines may result in personal injury or death or property damage. Proper set up entails following TI’s instructions for electrical ratings of interface circuits such as input, output and electrical loads.

 

[msis]

erm ... I am sorry but this is absolutely irrelevant @FvM

--- Updated Aug 8, 2020 ---

my TI synthesizer works brillantly btw (at least in the "low" frequency range that I don't need a mixer [frequency downconvert recieving structure] for)

The ADI interface is just as userfriendly as the TICS,

ADI senior engineering wrote an email where they try to hint at my power supplies being "faulty". Doing so, they write stuff which is 100% false (and any student in the first semester would notice that)

so they will have a chance to bring their own power supplies to the court hearing, if they dare to do so ... the devices thez sent me so far though are evidence enough

they have two options:
send me the devices as advertised or pay £ 10.000

 

[volker@muehlhaus]

So funny. Looking at the thread, you need an RF consultant instead of a laywer.

 

[msis]

I am really amused by now ..... at the hearing, we may just compare their famous 4.4 GHz synthesizer (deliberately defunctionalized) with the cheap Chinese spin off (£50 on banggood)

LOL ... any non expert judge will see the difference lol ... then they can go ahead and explain something about "negative dbm values" like they do in my email Loooooooool fucking idiots!

the result should be criminal proceedings if the judge is any good ... but certainly, he/ she will be able to see how it's supposed to look like on the spectrum analzyer and how the ADI device is not doing it

--- Updated Aug 8, 2020 ---

@volker. I have won 2 similar cases recently ... I don't need a lawyer for filing for £ 10..000. Is not necessary in the UK you know ... you can do that online ... takes about 15 min ..... I haven't tested all 5 ADI devices yet ... bc I have another deadline to take care of that's all ... it just gets a bit delayed ... but I told them officiall already when - at the latest / the filing will be made

 

[volker@muehlhaus]

So funny ... making all sorts of mistakes, no clue about RF levels, damaging devices by total ignorance to AC coupling, but blaming the component vendor.

You need to understand that these devices are designed for qualified users who know what they are doing. But if you can win a case like that ... congratulations, you might have found an innovative business model.