HL2 transverter companion card - hl2_xvtr

[Matthew]

This week I ordered the prototype PCBs for my hl2_xvtr project. My intent is to remove my N2ADR card and replace it with this PCB. A jumper cable from the HL2 PCB will be used to source the clock from the HL2 PCB external clocks. There are more clever things that can/could be done with multi-band switching etc. but I have taken a simple approach for now to satisfy my simple curiosity of experimenting on VHF. My jlpcb order should arrive in around 9 days. The gerber files I ordered are here:

https://github.com/m5evt/hl2_xvtr

My Mouser order has arrived already for the components. I am going to build a 6m variant first, but I would also like to test 2m and 70cm too. Supply of the SPF5189Z is/could be problematic. Mouser are expecting stock mid July. I was unable to find anything in the same footprint with similar performance. It is probably worth investigating an alternative in the future.

As you will see from the schematic there is nothing very novel about this. For people interested in duplex, OH2GAQ has already done excellent work on this and there wouldn't be much point in me re-inventing the wheel there.

The next stages are to design a suitable driver/amplifier (this will be a separate PCB). I will prototype a design with the PD85004, there are some good reference designs:

https://www.mariohellmich.de/projects/rf-amp_med-pwr/rf-amp_med-pwr.html

https://www.qsl.net/in3otd/ham_radio/PD85004_SE_PA/PD85004_SE_PA.html

http://www.bravoao.co.uk/g4bao/Files/G4BAO_70driver.pdf

However, I need to get my head around the S-parameters at lower VHF frequencies.

Once a working driver is confirmed, I will couple this to an MRF101 amplifier. The 50 MHz reference design in the MRF101 datasheet looks like a very good starting point. I am not looking to do a broadband design.

This is a prototype and unproven. But the gerbers and BOM are there if people feel confident soldering 0603 and want to order their own boards.

73 Matthew M5EVT.

[Steve Haynal]

Hi Matthew,

Thanks for your work and sharing!

73,

Steve

kf7o

[Matthew]

I have built up my 6m prototype and with my HF doublet antenna tuned to 6m, I've been decoding FT8 signals all over Europe and North Africa on 6m this AM. I've never had any capability to RX/TX on 6m before, in a short period this AM I can see why it is called the "magic band".

There are some modifications required to the current board. I've been updating issues in GitHub for the build. I will look to up-issue the PCB with these changes.

73 Matthew M5EVT.

[Christian Veith]

Hi Matthew,

Great news :)

The u.fl socket is unpopulated on the factory made PCBs of the HL2, isn't it?

Steve, would it make sense to populate it in fab for newer releases of the hl2?

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/5170a782-fe52-49bf-81bb-fec74071991en%40googlegroups.com.

[Steve Haynal]

Hi Matthew,

Thanks for sharing. It looks and sounds very interesting. I will ask if the u.fl scoket can be populated in the next run.

73,

Steve

kf7o

[radi...@mail.com]

Hi Matthew

I am following this with great interest. Early days I know, but wonder if you have given thought to switching between multiple transverters stacked in same case to make the HL2 into a multi-band VHF/UHF rig with separate antenna outputs of course? For myself I'd just be looking at the standards, so 6m, 4m, 2m, 70cms.

One HL2 for HF and another one for four band VHF/UHF would be great!

My other thought though........ I have linears on some of those bands and generally they take maybe 3W, maybe up to 10W to drive them fully. Any intention to have an onboard PA of any sort? I am guessing not and you will be thinking homebrew amps with low level drive? Not sure if I'm right, but have feeling most commercial linears do require higher drive levels than your intended 20dBm?

Do you have any performance benchmarks you are aiming for in terms of comparing to other transverters currently available?

Anyway, great project. I'm a newcomer to 6m myself and equally smitten. Check out Jim Bacon's fantastic propagation predictor website for live map showing potential Es paths. It's VERY accurate!

propquest.co.uk

(Click "EPI" tab for map).

73

Max

[Pierre Martel]

Very interesting project. 

I needed a reason to buy a second HL2. (not really but if I say it that way I feel less culpability) 

Pierre VE2PF

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/8b9986b7-0bac-44d5-bf34-f4cd94d4f115n%40googlegroups.com.

[Matthew]

To answer a few questions:

- I haven't tested 2m/70cm yet. I think I will build these up on the next rev of PCB.

- A driver amp within the case would be good. I don't think it would fit on the current PCB, I think it would have to be stacked. But then thermal management becomes quite important/tricky. In the first instance I am prototyping a separate driver module to see what is possible (1 W? 3 W?).

- Stacking these boards is a nice idea, but I'm not too sure how cost effective it is, it probably very quickly becomes more cost effective to buy an ADALM-PLUTO sdr. There would be a lot of component/functionality duplication across the stack. Plus, it would not fit in the HL2 case. It would probably be better served by a dedicated PCB that had all the 28 MHz IF components on, then switched filters etc. It would also seem a shame to have a stack and not have full duplex (satellites). Hopefully this project proves a few concepts as a stepping stone for such a project.

- Benchmarks - I don't really have any test equipment for > HF to quantify performance. It uses very low NF MMICs, I chose a +17 dBM mixer to hopefully have a reasonable IP3 figure, triple tuned BPF to block unwanted mixing products etc. Phase noise from CL2 should be reasonable, but I don't know what my buffer adds to this.

I think what would be most of interest to me to evolve this would be figuring out full duplex with RX (or TX) on 2m and TX (or RX) on 70cm for satellites using just CL2. I have no idea if I will ever find time to try this though.

73 Matthew M5EVT.

[radi...@mail.com]

Thanks Matthew for your response and all of the interesting and comprehensive answers. I have considered a Pluto for VHF/UHF. I'm never sure what the actual performance is like versus something like the HL2 plus a transverter? Also same issue with regard to the low output level. Would needs dedicated buffer amps to be able to add linears etc. To be fair I haven't done any research on that at all. I'm sure many have done it. Maybe I underestimate the performance of the Pluto? It's a chep option in many ways. I think lower end is 70 MHz so no good for 6m! But other bands, maybe.

Good luck with the project. I am still looking on to see where it goes!

73

Max

On Sunday, 18 July 2021 at 12:33:52 UTC+1 Matthew wrote:

To answer a few questions:

It probably very quickly becomes more cost effective to buy an ADALM-PLUTO sdr. There would be a lot of component/functionality duplication across the stack. Plus, it would not fit in the HL2 case. It would probably be better served by a dedicated PCB that had all the 28 MHz IF components on, then switched filters etc. It would also seem a shame to have a stack and not have full duplex (satellites). Hopefully this project proves a few concepts as a stepping stone for such a project.

[Brian Holmes]

This looks very interesting. 

Brian

VE3IK 

On Jul 18, 2021, at 10:43 AM, radi...@mail.com <radi...@mail.com> wrote:

Thanks Matthew for your response and all of the interesting and comprehensive answers. I have considered a Pluto for VHF/UHF. I'm never sure what the actual performance is like versus something like the HL2 plus a transverter? Also same issue with regard to the low output level. Would needs dedicated buffer amps to be able to add linears etc. To be fair I haven't done any research on that at all. I'm sure many have done it. Maybe I underestimate the performance of the Pluto? It's a chep option in many ways. I think lower end is 70 MHz so no good for 6m! But other bands, maybe.

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/6b6b12fa-2f6b-42bc-8278-ef4f3b594c8dn%40googlegroups.com.

[Matthew]

I am looking to up-issue this PCB. The question has been raised asking if this could sit concurrently with an n2adr board installed in the case.

If we work under the assumption that in the HL2 J1 is fitted (I've removed it, but that is to suit my needs), the problem that has to be overcome is any signal on the n2adr rx line will be present in parallel with the RF3 (hl2_xvtr RX signal) signals. So the relay K2 would need to be in TX to guarantee we only had hl2_xvtr signals making it through to the ADC (regardless of the xvtr IF frequency used). I don't think the gateware currently supports this, but I imagine it wouldn't be too tricky to implement.

However, my intention had been to keep the option open to stack a driver PA into the housing above/below this xvtr card. I don't think this is mechanically/thermally possible with an N2ADR filter also installed. Then if I wanted to operate on 2m instead of 6m, I would quickly change out the hl2_xvtr card (I've never actually had the top on my HL2 case so this is a quick operation for me).

What is the preference for those who want a solution within their existing HL2 housing? More than 17 dBm out and no n2adr filter, or concurrent operation with n2adr board and only 17 dBm out (and likely PC software support/modification required to enable this configuration)?

Of course if the user housed this companion card in a separate box, this makes topic somewhat less of a concern.

73 Matthew M5EVT.

On Sunday, 18 July 2021 at 12:33:52 UTC+1 Matthew wrote:

[Josh Logan]

My vote, more than 17 dBm out.  I would rather plan on 2 HL2s one for HF and one for transceiver use.

Use the 55mm case and see If you could make it switch 3 bands/modules.  Start with 6m, 2m, and 70cm, and then possibly offer 3 more plugin modules for 1.25cm, 33cm, 23cm.  Turn it into a modern FT-736R where you can pick which modules to load.

73, KD7HGL

Josh

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/083780ec-a3ca-4d1e-96ef-1701337d9952n%40googlegroups.com.

[radi...@mail.com]

Really not at all sure. Started out thinking maybe more than 17 dBm but then the appeal of several switchable/stackable cards whilst retaining HF capability is also appealing. I guess the only issue is lack of commercial VHF/UHF linears that can be driven by 17dBm output? Difficult choice!

73

Max

[David Wilde M0WID]

My vote would be to have both n2adr and transvertor boards in the case.  17dbm out is fine for me, with external PA.  Stackable transvertor boards would be even better, but not a must have.

Great projects Matthew, I would be interested in trying a 70cm version.

73

Dave M0WID

[Walter Holmes]

Very well done there Matt.

I too would like to see this as stackable boards, so that N2ADR and transverter boards were in the same case.

But if that's too tight to fit in the same case, a matching case with the transverter board would be a perfectly fine solution.

And if Maker Fabs were able to crank these out already assembled, that would certainly be the best of all worlds.

Several of us have more than 1 Hermes, so if the system could not support both the N2ADR and transverter, the HL2 is certainly cheap enough to have 1 for each band. :) 

Al the best, and we can't wait to see how far your able to carry this project. Keep up the GREAT work on this.

Walter/K5WH

[k...@khk.net]

My vote is for a combined HF/VHF/UHF setup. I’d like to use the HF bands most of the time, but still be able to get on 6m and 2m. 

Thanks,

KarlHeinz-K5KHK 

[Matthew]

A short update.

I have built version 2 of this PCB for both 2m and 6m. I have confirmed the issues from version 1 are fixed. I can hear 6m FT8 signals and a local 2m repeater ID every 15 minutes. Apart from that 2m and 6m are quiet...

The RSGB VHF contest calendar shows that there are some contests in the first week of November on 2m. I will try and homebrew a beam antenna and have a listen to confirm that I can hear some longer distance signals.

The gerbers for v2 are on my GitHub page. I will try and find some time at the weekend to update the BoM for the 2m variant.

73 Matthew M5EVT.

[Bill Cox]

This looks awesome.  I'd love to build one and replace my Ukrainian xvtr if I could drive enough power.  I prefer to have a dedicated receiver per antenna, so replacing the N2ADR spot works well for me. 

Is the xvtr any taller than the N2ADR?  I could fit an interface board over it if it is not taller.

I'd be interested in having more power than 1W.  Entering my transfer is 10 watts but it doesn't seem to put out more power than my three watt HT.  Would it be practical to put a higher power amp in the case?  E.g. 5w is in line with existing thermals.

73 Bill ak3q

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/d083f883-23f9-421f-aae5-b29b42a3b2een%40googlegroups.com.

[Matthew]

Hi Bill,

The tallest component on the PCB is the phono jack (same as the one on the N2ADR filter board).

I'm not too knowledgeable about VHF amplifier design I'm afraid. I suspect it could be possible to put something in the enclosure. However, for now I am pursuing an amplifier based on the MRF101 dev kit. According to the datasheet for the reference design, I should be able to get 80 W out on 6m driving it just from this board. There is also a reference design for 2m too. The dev kit is cheap and you even get a spare FET. The only down side is a 48-50 V supply is needed, but they are easy enough to find once you start looking.

73 Matthew M5EVT.

[Matthew]

I've updated my GitHub pages with the BoM for a 2m variant. The 2m component values are text on the schematic, so just beware if you are trying to do anything clever with auto generating a BoM from the KiCad files (and you are interested in the 2m variant).

https://github.com/m5evt/hl2_xvtr

I have 2 spare PCBs. If anyone is interested in experimenting with building one of these and is comfortable with 0603 size components, please get in touch. Alternatively, the gerbers are available on my GitHub page and you could order 5 PCBs just by uploading the gerbers.zip file to jlpcb.

73 Matthew M5EVT.

[Michael Durkin]

I'm comfortable in soldering SMD .... I would suggest an AMP for VHF with a RA80H1415M1 ...  and control module like this from Ebay,  284403066977 -- i have a hard time finding it elsewhere

KC7NOA

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/0dc01210-2917-4364-8d12-c38955581e03n%40googlegroups.com.

[Mike Lewis]

For the PCB, W6PQL has a similar board without the relays that might work.   You can find it under his “Parts I can Supply” nav link.   I use a few for 1296 modules as described here.

 

Mosfet Module Amplifier Construction Tips (w6pql.com)

 

I have seen similar boards on eBay from time to time. Depends on when you look. 

 

 

Mike

K7MDL  EL87sm & CN88sf

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAC8M0Esa137QOHsJ_UzeR8cjpjqR35bvxsRFMtBeRZGJcy9c_w%40mail.gmail.com.

[John Williams]

I would like to point out this article... I ordered some of these boards from Oshpark. Cheap to buy. Maybe a good way to try a prototype...

https://www.qsl.net/in3otd/ham_radio/PD85004_PP_PA_II/PD85004_PP_PA_II.html

I have a few boards to share. Or I can upload it to Oshpark...

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CO6PR18MB385897DF8EC40A9990C0502AF6869%40CO6PR18MB3858.namprd18.prod.outlook.com.

[Heikki OH2LZI]

Hello group,

I have been following the HL2 transverter story and might consider building one this winter. I will, however, make my own design both mechanically and electrically.  My existing transverter (abt 10 yrs old) is based on discrete components but now I find the MMIC amps quite attractive. It seems, however, that those listed in the schematic are more or less obsolete. In my understanding the one in the RX frontend should have low noise figure while for the others NF is not that critical. My question is: has anyone studied the existing (not obsolete !) amps that might fit in this design. In my case the footprint is  not an issue. 

73 de Heikki (OH2LZI)

[Matthew]

Hi Heikki,

The PGA103+ is likely a suitable replacement and also cheaper.

73 Matthew M5EVT.

[Heikki OH2LZI]

Hi Matthew ,

Tks for the info regarding the transverter !

I understand that You have  made some experiments also with an MRF101 amp at 50 MHz. Any results ? I would be  interested in building a narrowband amp for 6 m to boost up my transverter with 4w output. The reference design in the MRF101 datasheet looks OK for that but I wonder if it is linear or class C  ?  

73 de Heikki (OH2LZI)

[Matthew]

I've not advanced to a point where I can measure linearity yet.

If you haven't seen it, this project might be of interest:

https://www.la1k.no/2020/01/22/afm907n-a-10-w-hf-amplifier-with-caveats/

73 Matthew M5EVT.

[Paul Loughnane]

Matthew, I'm planning to have a go at building this in the near future. I'm also happy to experiment so wondered if what you have is already a solid starting point, or there are any improvements you can already recommend based on your work to date. As mentioned above I'll likely substitute the PGA103+ in the front-end as the SPF5189Z does appear to be obsolete, or maybe the  PSA4-5043+ as speced in the Anglian 3L? 

[Matthew]

The files on github were my second revision of this PCB. I have built, tested and used a 2m and 6m version of this second revision. There are no improvements/mods that I recommend making for this revision. I look forward to hearing your experiences with this.

73 Matthew M5EVT.

[Heikki OH2LZI]

Just to inform that I have managed to build a 50 MHz MFR101 narrowband amp based on the schematics in the NXP data sheet.  Currently the max output  is only 35-40W  limited by my power source of 2A/50V. This consists of a 120W computer power supply (20V) and a Chinese voltage booster up to 60V. I have followed the NXP schematic quite closely, only the coils are homemade but measured with NanoVNA. It is my feeling though , that more output should be available and the circuitry may need some more tuning or even redesign. Looking forward to results from other members !

73 de Heikki (OH2LZI)

[Heikki OH2LZI]

Hi group,

Interested if anyone has been more successful with the MRF101 amp circuitry at 50 MHz ? According to NXP 100W output should be easily available with minimal drive (< 1 W) but in my case something happens at 30-35 W out/ 4 W in and the output  nor the supply current will not increase anymore. I wonder if the values given for output matching L and C are optimal ?

73 de Heikki (OH2LZI

[Matthew]

Hi Heikki,

I have built the circuit but got side tracked by some other projects and never installed it on a proper heatsink to do high power tests.

I've just ordered a big heatsink on ebay so perhaps I can get this tested (and report back here) before I miss another sporadic E season on 6m.

73 Matthew M5EVT

[Heikki OH2LZI]

Hi Matthew,

I fully understand,  several projects going on simultaneously here. too ... Trying to prepare for the upcoming 6-m season also with a new SWR-meter plus a Z-match. The output from MRF101 amp, 30W at 50MHz,  is not that bad, 5 - 10 dB more than previously, but a full 100W (or even close to that) would be appreciated !

73 de Heikki (OH2LZI) 

[Heikki OH2LZI]

Hi Matthew et al,

Any progress with MRF101 amp for 6 m-band ?

73 de Heikki (OH2LZI)