Transverter Question

[Bill M]

I am thinking of buying a used 2m transverter.  It has a split IF. That is there are separate connections for RX and TX between the transverter and the transciever.  So, on a hl2b6 with the N2ADR filter board can I connect TX to the low power output, turn off the Hermes power amp, and hook up RX to the main RX/TX connection for RX?  Now that I'm thinking about it is a transverter w/split IF required if using the low power output? It would obviously be if the low power output is output only.  Perhaps the RX only connection (not accessible w/the N2ADR board) is needed? 

73,

Bill w8ei

[neil whiting]

Hi Bill,

I have my HL2 connected with the low power output via a 16dB attenuator to the transverter TX input (needs to be 1mW as also used with a K3).

I chose to add another SMA connector to the HL2 for the RX signal from the transverter. This is wired using thin coax to the alternate RX input,

which can be accessed on the N2ADR board or (as I chose to do) at the connector between the boards.

Doing it this way means the transverter will never get 5W up its RX output even if I accidentally enable the PA.

Good luck with your project.

73, 

Neil  G4BRK

[terry long]

Hello Neil,

Thanks for the transporter information.  Can you take some photos of your setup and post for the group?

73,

Terry N8AB

[Steve Haynal]

Hi Terry,

The HL2 is designed such that you can disable the PA and disable the TR. This allows for full duplex operation even with the N2ADR filterboard. Low power output is TX, usually switched TX/RX becomes RX. The danger is that you can apply full power to the TX/RX if you don't have things set properly. 

73,

Steve

kf7o

[neil whiting]

Hi Terry,

Sorry for the very long delay, finally got around to it.

Picture showing transverter RX connection to the HL2 board header.

73,  Neil  G4BRK

[Steve Haynal]

Hi Neil,

Thanks for the picture of the mod. It looks very useful.

73,

Steve

kf7o

[Max]

I am very interested in using one or more transverters with HL2 (v9) to add VHF/UHF capability. But I want to be able to quickly change band as easily as possible, and also to not impair the normal HF operability.

One possible candidate is this one, which is aimed at Flex users but I think should also play nicely with HL2 with the "RX In" mod that Neil has described in his post on this thread.

https://www.ebay.co.uk/itm/254510549922

To be clear, I am not looking for the cutting edge of performance (yes I am aware of Kuhne etc.)...... just usable performance to have some fun on some other bands. I am aware of the possible limitations of this item in terms of top level  performance, which is not what I need for my casual use.

So a lot (sorry) of questions from  a "layman" as I am often left with much head scratching from the majority of posts on this board...... I am aware my technical knowledge is many levels below the average conversation on here. And yes, I did look for the answers by searching previous posts but just could not dig out enough simply understandable information for definitive answers to my simple questions.

Q1: I want to use separate RX Input instead of main HL2 Ant In/Out connection because (I agree with Neil) I know for sure that if I use the latter as my main transverter receive input I will very soon fry the transverter by having the HL2 PA enabled accidentally. I cannot find the relevant pins for Neil's mod on the N2ADR header/HL2 header documented in the Wiki. Have I missed it? Any chance of some info on how to implement this option? Do I need to mod the HL2 board or just connect mini-coax to the header like Neil shows? I looked on HL2 schematic but I don't see the info. Must be missing it? Does using this option affect the normal HF operation with N2ADR board?

Q2: Does using the separate "RX in" load the RX in some way if the transverter is turned off or not in use when I want to use normal HF mode on HL2? Or does the transverter need to be detached from the RX connection when not in use? Or is this a totally separate "RX in" path that is switched in software when in use? I exclusively use SDR Console as my software. I don't see any option to switch to this input in the software? Presumably it is needed, as I will still have my HF antenna attached to the main N2ADR TX/RX antenna input. Maybe I need to request Simon to add this feature?

Q3: Obviously with this particular transverter I am looking at using using low level RF1 out to drive the transverter. Can this be adjusted down to 0 dBm output level? If so, how? In the host software? I don't see any control for adjustment of this output level in SDRC? Or is it always preset to 17 dBm in Gateware and I need to use an attenuator to pad it down to 0 dBm?

Q4: I guess low power output RF1 is "raw" from the HL2 with no filtering. What are the implications of this for a transverter? With a conventional transceiver, I guess the harmonic outputs are limited by low pass filtering at the output of the TX, but with SDR like HL2 here there will be none when RF1 is used instead of the main filter board output. Should there be some sort of filtering applied between RF1 and the transverter input? Would it be better to use a different model of transverter that takes 5W drive instead of low level so that the benefit of the N2ADR board is added? Perhaps RF1 output relies on the transverter having decent filters on input and output? Or is it inherent in the design of any well designed transverter that the necessary filtering is present?

Q5: With PA disabled on HL2, is this adequate to prevent very low power output on 28MHz radiating from the main antenna output, or perhaps main HF antenna needs to be disconnected from antenna output when transverter is used?

Thanks for looking.

73

Max

[Steve Haynal]

Hi Max,

Neil is using the RF3 input on the rear IO strip as seen here. On the schematic sheet 5 of 7 you can see that RF3 is permanently connected (not switched) to the ADC input. Receive degradation can occur if there is a 50Ohm input at RF3 and 50Ohm input via RF1 and the TR relay. There are several ways to avoid this:

** Use high impedance as done with the PureSignal attenuators. This reduces RX degradation.

** Don't connect anything to RF1 (high impedance open circuit) and use low power TX on RF2 and RX on RF3. This is the intended transverter model. If you still want to connect something to RF1, then we would have to add support in the gateware to keep the TR relay always in TX but no PA on. But when you go back to standard operation, you will still have RF 50Ohm on RF1 and RF3. 

** Add external relays to that a 50Ohm receive load is only connected to RF2 or RF3 depending on whether you are using transverter or standard.

There is an example board for PureSignal feedback here.

See additional comments embedded below.

73,

Steve

kf7o

On Thursday, June 25, 2020 at 3:44:54 PM UTC-7, Max wrote:

I am very interested in using one or more transverters with HL2 (v9) to add VHF/UHF capability. But I want to be able to quickly change band as easily as possible, and also to not impair the normal HF operability.

One possible candidate is this one, which is aimed at Flex users but I think should also play nicely with HL2 with the "RX In" mod that Neil has described in his post on this thread.

https://www.ebay.co.uk/itm/254510549922

To be clear, I am not looking for the cutting edge of performance (yes I am aware of Kuhne etc.)...... just usable performance to have some fun on some other bands. I am aware of the possible limitations of this item in terms of top level  performance, which is not what I need for my casual use.

So a lot (sorry) of questions from  a "layman" as I am often left with much head scratching from the majority of posts on this board...... I am aware my technical knowledge is many levels below the average conversation on here. And yes, I did look for the answers by searching previous posts but just could not dig out enough simply understandable information for definitive answers to my simple questions.

Q1: I want to use separate RX Input instead of main HL2 Ant In/Out connection because (I agree with Neil) I know for sure that if I use the latter as my main transverter receive input I will very soon fry the transverter by having the HL2 PA enabled accidentally. I cannot find the relevant pins for Neil's mod on the N2ADR header/HL2 header documented in the Wiki. Have I missed it? Any chance of some info on how to implement this option? Do I need to mod the HL2 board or just connect mini-coax to the header like Neil shows? I looked on HL2 schematic but I don't see the info. Must be missing it? Does using this option affect the normal HF operation with N2ADR board?

To support both transverter and standard operation, you must make sure that only 1 RF load is present on RF3 or RF1 during receive. The HL2 can be in transverter mode or standard mode but requires manually switching RX antennas. You'd have to add at least one relay to support both automatically and not affect the normal HF operation with N2ADR board. 

Q2: Does using the separate "RX in" load the RX in some way if the transverter is turned off or not in use when I want to use normal HF mode on HL2? Or does the transverter need to be detached from the RX connection when not in use? Or is this a totally separate "RX in" path that is switched in software when in use? I exclusively use SDR Console as my software. I don't see any option to switch to this input in the software? Presumably it is needed, as I will still have my HF antenna attached to the main N2ADR TX/RX antenna input. Maybe I need to request Simon to add this feature?

RF3 is not switched. You would have to add a new relay. You could possibly use the P13 or bit 7 output from the N2ADR board.

 

Q3: Obviously with this particular transverter I am looking at using using low level RF1 out to drive the transverter. Can this be adjusted down to 0 dBm output level? If so, how? In the host software? I don't see any control for adjustment of this output level in SDRC? Or is it always preset to 17 dBm in Gateware and I need to use an attenuator to pad it down to 0 dBm?

The AD9866 supports 7.5dBm variation of the TX output. You can lower the output to about 9.5dBm. You can use software to lower that more via decreasing the amplitude of the signal at the expense of data bits. There are modes to the PA preamp which can lower power more, but those also lower the PA output power.

 

Q4: I guess low power output RF1 is "raw" from the HL2 with no filtering. What are the implications of this for a transverter? With a conventional transceiver, I guess the harmonic outputs are limited by low pass filtering at the output of the TX, but with SDR like HL2 here there will be none when RF1 is used instead of the main filter board output. Should there be some sort of filtering applied between RF1 and the transverter input? Would it be better to use a different model of transverter that takes 5W drive instead of low level so that the benefit of the N2ADR board is added? Perhaps RF1 output relies on the transverter having decent filters on input and output? Or is it inherent in the design of any well designed transverter that the necessary filtering is present?

The RF1 is from the PA preamp, which is a pretty rugged op amp. The sensitive AD9866 is protected. The HL2 has a LPF (~32 MHz and lower) on all transmit paths. I'm not a tranverter expert, so will pass on the other questions. Take a look at the document by Hamish for more transverter ideas:

https://groups.google.com/d/msg/hermes-lite/yBw-FH3ir0I/TUICkzcjBgAJ

 

Q5: With PA disabled on HL2, is this adequate to prevent very low power output on 28MHz radiating from the main antenna output, or perhaps main HF antenna needs to be disconnected from antenna output when transverter is used?

I think the relay and PE4259 isolation is good enough that you won't see meaningful power on 28MHz. 

[Alan Hopper]

Hi Steve and Max,

perfect timing, I had been wondering about this as I have this https://www.ebay.co.uk/itm/221794654735 on order.  My local club has been holding a daily 2m covid net (2m is the distance we are meant to keep apart in the UK) and at the moment I have to climb the hill I live on and use my baofeng handheld. My first thought was to use the lime sdr but this seemed cheaper and simpler in the end. I'd be interested to hear if anyone has any experience of these transverters. I didn't put much research into this purchase as the price was so tempting and it at least gives me something to test transverter use with spark.  

73 Alan M0NNB 

[Marc OLANIE]

Hi Alan

I’m afraid you’ll probably regret this « false bargain »

The « N2 » version needs eavy modification to obtain less than average working conditions, and the « N1 » release is definitely unusable .

If you need more technical infos, you’ll have to cut&paste&googletranslate this french written analysis

http://www.radioamateurs-france.fr/wp-content/uploads/Transverter-28_144_432-MHz-UT5JCW.pdf

 

Same remarks concerning the dual band transverter version sold on ebay, which is a kind of adaptation of the former 28/144 kit. Twice the price, half the already poor performances and twice the design errors

 

I’m still working on the OH2GAQ set. So far, it is probably the best answer to the different questions asked by Max. The input and output filters should perhaps need some small modifications to offer a nice 10 MHz bandwidth (20 to 31 MHz spectrum at the I.F. level) , but it’s a detail.

 

VY 73’

Marc f6itu

 

 

De : 'Alan Hopper' via Hermes-Lite <herme...@googlegroups.com>
Envoyé : vendredi 26 juin 2020 07:59
À : Hermes-Lite <herme...@googlegroups.com>
Objet : Re: Transverter Question

--
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/7f4c630d-3f00-4adb-84ae-7354a3003a1bo%40googlegroups.com.

[Alan Hopper]

Hi Marc

thanks for that, I was not expecting much :) Sometime these cheap purchases are very useful in working out what you really want and need, just like building or writing a prototype.

On another track I've been wondering about coherent reception at vhf and uhf as the antennas are more manageable, I've not really thought this through but I'm guessing the xvertrs will need to share a clock.  I'm very interested in Hamish's solution and may ultimately go that way but for now step one is chatting to the locals.

73 Alan M0NNB

To unsubscribe from this group and stop receiving emails from it, send an email to herme...@googlegroups.com.

[neil whiting]

Sorry guys, having a lot of finger trouble these past 3 days - I blame the 30deg heat.

Try again.

Hi Max,

A great set of questions.

Steve has answered a number of them.

For your Q4 and Q5, I added a little module external to the HL2 which takes the low power output from RF2 through a 28MHz bandpass filter

and an attenuator to get the level down to 0dBm. The filtering may be overkill as I didn't do any measurements to show if it was needed or not.

Transverters do not generally provide good filtering on the IF but the signal frequency filtering will help with further-out unwanted spurii.

I can't remember quite what filter design I used and it is soldered into an FR4 box so I can't give you a picture easily. I am trying to do a better job

of documenting my projects - something i should have started many years ago

Something like a 10m one of these https://www.qrp-labs.com/bpfkit.html might do the job.

There is room inside the HL2 with the taller case to fit the filter/attenuator inside and I may do that in the future.

I have arranged it so I can easily unplug the HF antenna when I am using transverter mode, but in practice I haven't noticed any 10m breakthrough.

Maybe that will change as the sunspots return.

One further convenience for me would be the automatic switching of transverters (I have 8 of them). I can do this from the K3 which provides 4 bits

from which to decode the transverter band. In theory it should be possible to provide these from the HL2. Quisk provides transverter support very nicely

with local oscillator offsets for each band. The HF bands are decoded to switch the N2ADR board filters and this is configurable within Quisk.

The difficulty is to make it switchable between transverter and normal mode. If another I2C interface was added it could provide transverter band

select signals, but then another bit of code and configuration would be needed to control this. I'm hoping to look at doing this but other projects

keep getting in the way.

73,  Neil  G4BRK

[Max]

Steve

Many thanks for your reply. That makes things much clearer. For some reason I thought the IO strip was referring to and end panel, not the main IO rear HL2 board and now I clearly see RF3. I am the dummy and also need to pay more attention! Also, not seen reference to RF2 before. That adds a bit more confusion for me but I will read and analyse carefully. I thought RF1 was the low power output? Anyway, some switching differences here obviously so it seems so I will read and inwardly digest.

Thanks again and 73

Max

[Max]

Alan & Marc

I am no expert (very far from it) but I did read as many reports as I could find about these Ukrainian transverters. I hope I don't have my facts wrong, but it is claimed elsewhere that original transverter design is by UR3LMZ who trades as hfvhfparts on eBay. I don't think he has his own website. Anyway, he now produces what he calls "HD" versions of these transverters, the difference being (as far as I can see) that he has upgraded the mixer to Minicircuits ADE DBM chips. He says the transverters-store are copies of his original designs. Not saying it is true, just repeating what he has said. As far as I can see the transverters-store items do not yet use the ADE mixer. Marc...... I don't now if this is likely to offer a significant improvement to the performance?

For me so long as I can produce a legally acceptable signal without a nuisance to others and have some fun then that's about all I need, although obviously I do want to obtain an acceptable standard of performance.

See here for HD versions from UR3LMZ:

https://www.ebay.co.uk/itm/132530907793

https://www.ebay.co.uk/itm/132326955863

and others:

https://www.ebay.co.uk/str/hfvhfparts?_trksid=p2047675.l2563

I might actually go down the board-only route and build in matching cases to the HL2.

Marc, I did read your report. Very interesting. Why do you say the dual band versions have half the performance? Is the performance not exactly as per the single band versions?

Also sadly I am very time poor so I don't have time for a scratch built project these days. Just want to get on air as soon as possible, so ready built modules have huge appeal. If only all projects were of the same standard as HL2!

73

Max 

[Max]

Many thanks Neil. More very useful information. Glad I asked the questions as I can see much clearer now how I will want to proceed.

73 to all

Max

[Max]

Apologies Steve, now seeing all on schematic. Of course RF2 is main ANT input/output. My bad. I really am the dummy. I know if I use this I will have a destroyed transverter in about 2 days, so pretty sure that rules it out for me.

73

Max

[Charles Brookson]

Probably your best bet is to build a kit from Sam DDK like his 70cm Iceni kit

http://www.g4ddk.com/Products.html

It has pretty good performance, and is well priced. I have the 4m kit which I'm planning to drive with the hermes-lite.

73s

Charles G4GBA

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/2d21ad4f-937c-42ac-8379-4fd73a8b4f6co%40googlegroups.com.

--

LargePrefPlaceholder-XKUz1MEJBwkOM

[Max]

I must be missing something Charles. I don't see any transverter kits for sale? And also no info about anything other than 70cms transverter. I want 50,70,144,432. That's also a lot of building for me that I don't think I would ever get time for sadly.

Not writing it off though. Be nice to see the product. I note on his QRZ.com he mentions transverter kits for 70,144,432. Shame no 50MHz as that's the one I am most interested in coming up to the E's season.

I will drop Sam a line and see what he says. Thanks for the tip.

73

Max

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/2d21ad4f-937c-42ac-8379-4fd73a8b4f6co%40googlegroups.com.

--

LargePrefPlaceholder-XKUz1MEJBwkOM

[Steve Haynal]

Hi Max,

I think I used the wrong labels for the external connections. Here is a summary:

RF1 - Low power output. Enabled by disabling the PA. See the videos. Labeled RF1 on end plates.

RF2 - Relay switched main output labeled ANT on end plates. Either 5W out or RX in. Gateware already has option to keep the relay in RX if RF1 is used for low power output to provide full duplex. Gateware can be enhanced to keep the relay in TX if RX from RF3 is desired.

RF3 - Alternate RX in, not switched.

Any configuration should be possible, the the game is to not have RX loads on RF2 and RF3  at the same time. 

73,

Steve

kf7o

[Max]

Many thanks Steve. Appreciate the summary and clarification. Useful.

73

Max

On Friday, 26 June 2020 22:07:04 UTC+1, Steve Haynal wrote:

Hi Max,

[Ronald Nicholson]

Hi Steve,

What do these various external RF connection modes mean in terms of control bits inside the Metis/UDP packets?

Thanks,

Ron

n6ywu

------

On Friday, June 26, 2020 at 2:07:04 PM UTC-7, Steve Haynal wrote:

Hi Max,

[Max]

Following on from my earlier question, I am now looking at constructing a switch box to connect four transverters to HL2 using RF1 and RF3 for the transverters and retain RF2 for output to HF linear. I am actually tempted to put all four transverters and the switcher into one case, and only activate the transverter in use.

Thinking of using this to switch the 28 Mhz inputs/outputs:

https://uk.farnell.com/skyworks-solutions/as179-92lf/ic-switch-rf-spdt-0-02-3-0ghz/dp/1753762 

I was going to use these but they seem to have gone out of stock UK and now seem very expensive in the USA ($9 a piece.... phew):

https://www.minicircuits.com/pdfs/MSWA-2-20.pdf

Got idea from G6PQL here:

https://www.w6pql.com/automatic_transverter_interface.htm

Anyone have any thoughts on the substitution of he Skyworks part? OK for 50 ohm impedance or not really an issue at 28 MHz?

Thanks for looking.

Max

[Steve Haynal]

Hi Ron,

Currently, the only bits in the protocol affecting this are:

0x09	[19]	Onboard PA (0=off, 1=on)
0x09	[18]	Q5 switch external PTT in low power mode or 0=ATU tune and 1=ATU bypass when PA is on

When the onboard PA is off, you can set addr 0x09 bit 18 on so that internal PTT switch (HL2 TR relay) still switches between TX/RX. (There is a typo on the protocol page, external should be internal.) The external PTT switch (EXTTR on the back panel) always switches. The main reason for this was so that the ANT connection could be used for RX (not switched to TX) while the low power RF1 was used for TX. This essentially provides a full duplex solution with the existing 2 RF connectors. Since we had to have a bit to disable TR switching when RF1 was in use, I made it configurable via the protocol. I am not sure of any good reason why you'd want to switch the internal PTT when RF1 is in use unless you have removed the HL2 TR and have another switch somewhere else. I am open to removing or changing this behavior.

Another interesting possibility is to keep the internal TR in TX mode but PA disabled when low power RF1 is in use. This would allow full duplex via RF1 and RF3. The TR in TX mode would remove any additional load from the receive traces when RF3 is used for RX. But there would have to be an additional means to remove RX load from RF3 when you want to go back to using RF2 (main ANT with TR switch) for RX.

73,

Steve

kf7o

[Alan Hopper]

Hi Max, 

another approach to switching transverters is just to have a HL for each one. This is not the cheapest solution but you only need the HL2 with no filter board and you avoid all rf switching so it might not be as extreme as it first sounds. It would also allow running on multiple bands at the same time and possibly save on expensive coax by allowing the radios to be placed nearer each antenna.

If the HL nano is ever born this might then become the afforable way to go.

73 Alan M0NNB

[Hamish Kellock OH2GAQ]

Hi Alan,Max and others,

I have been using an interface system for some time with HL2's for interfacing to several microwave transverters. I have a full-duplex interface between the HL2 and a specially designed 2M (full duplex, separate Rx and Tx mixers) transverter, which can be switched to up to 7 higher frequency microwave transverters with 2M IF's. 6 of these can be traditional "half duplex" transverters, 1 can be a full-duplex transverter (for example for QO-100 operation). I use the separate lo-power output of the HL-2 and the separate RX input of the HL2. I have 20 to 31 MHz BPF's in the 2M transverters, so no separate filters needed after the HL2. I also have a separate interface card with I2C interface, and Arduino Micro on board, this allows all special functions to be done without changing the standard behaviour of the HL2. I disable the 5W PA on the HL2 by removing the output transformer, so even if I misconfigure something there shouldn't be a problem (famous last words !) A few other hams are in the process of building the same system, lets see how it goes. I posted this material on my website, it includes some notes, gerbers for the PCB's and simple SW for the Arduino. Have been using Quisk exclusively with R Pi 4's, but it should be OK with other SW. There are also some performance figures for the system.

link:   https://www.qsl.net/oh2gaq/

See the projects, Full Duplex MTC pages.

73,

Hamish, OH2GAQ

[Duncan Clark]

In message <37ed6680-6495-4669...@googlegroups.com>,
Hamish Kellock OH2GAQ writes

> A few other hams are in the process of building the same system, lets
>see how it goes.

Hi Hamish,

I am building a few of your transverter boards here.

I have re-done the filters so I now have 6m, 4m and 2m for full duplex.

I am using 7mm Unicoil Coilcraft filters (Mouser) with shielded ones for
2m and 4m and unshielded for 6m and T37/T50 toroids for the 28-31MHz
bandpass filter. Filters tuned up perfectly without any MMIC's in place.

Junk box +17dBm SRA-1H's for the mixers on Rx and some Ebay sourced
Australian RFLMX-1-13 +13dBm ones for Tx (cheap!)

ERA2 MMIC for the MSA2111.

I have a 20MHz OCXO in the junk box so I can use that with 30-31Mhz on
HL2 for 6m and possibly double the OCXO for 4m. I will worry about 2m in
due course as I have no antenna for that band.

My current issue is finding any G5Y-1 relays. Obsolete. I found some on
Aliexpress but they are all pulls and don't work :-( Annoying!

Anyone know of any available alternative with the same footprint to the
OMRON G5Y-1 12V DC (DIP format).

Duncan


--
Duncan Clark
G4ELJ

[Marc OLANIE]

Hi Duncan

I bought some of these relays on Aliexpress and had a lot of chance to get some working devices.
I even confirmed it to Hamish... but I only tested 3 of them

All of these relays are used, desoldered without precaution... and indeed, I had two or three dead fish in the tank. But you're right, a reliable second source would be great. I was even considering making a new drawing of the board with a different relay footprint.

Btw, could you pse tell me the shielded Coilcraft reference you used for the 2 m band ? I'm really interested

Cheers
Marc f6itu


-----Message d'origine-----
De : herme...@googlegroups.com <herme...@googlegroups.com> De la part de Duncan Clark
Envoyé : mardi 30 juin 2020 21:26
À : herme...@googlegroups.com
Objet : Re: Transverter Question

--
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/YouJDyI4G5%2BeFwij%40ntlworld.com.

[Duncan Clark]

In message
<BYAPR01MB5016E19B20...@BYAPR01MB5016.prod.exchangelabs.c
om>, Marc OLANIE writes

>Btw, could you pse tell me the shielded Coilcraft reference you used
>for the 2 m band ? I'm really interested

2m 146-06JO8SL
4m 146-10-JO8SL
6m 146-09JO8L

4m capacitors

Series ones: 6.8pf, 1pf, 1pf, 6.8pf
Parallel ones: - by design 9.4pf, 15pf, 9.4pf -

However I had to tweak these for less C i.e. 8.2, 11, 8.2. - didn't
write the values down but originally too much L when I came to tune.
Even now ferrite is maybe 1mm out of coil.

6m capacitors

Series ones: 8.2pf, 1.2pf, 1.2pf, 8.2pf
Parallel ones:15pf, 22pf, 15pf

The two inductor input bandpass filters were the same coils and
capacitor values - you can work out which from the schematic.

For a 20MHz LPF I used 220nH SMD inductors and capacitors of 160, 220
and 160pf.

For a 40MHz LPF I used 330nH SMD inductors and capacitors of 430 (330 +
130), 560 and 430pf.

Both these filters start dropping at maybe 1Mhz before the crystal
frequency and are down maybe 1.2dB at 20Mhz and 40Mhz but are also >40dB
at 2x frequency. I simply picked the closest L's I could buy and matched
the C's without any tweaking them on the board itself

Filters all designed using ELSIE but the LPFs' simply use preferred
values. I figured 1dB additional loss to be irrelevant when I have
plenty of gain to spare.

On the relays I bought 2 x 10 but have only 4 that work. Most open
circuit on the relay coil. Only received them today but I will be
complaining.

Hmm just looked on a shelf and I actually have several NAIS ARE134H 12V
SHF relays with the same footprint - ridiculously cheap Ebay buy a long
time ago that I had forgotten about. They are 75R but IMHO so what for
what we are using them for.

Even now on Ebay you can get 100 for £27.77 plus shipping of £17.85 plus
of course any import tax etc .

[Hamish Kellock OH2GAQ]

Hi Duncan and Marc,

I didn't realize this relay choice would be such a problem, originally designed the transverter using components that I had. I am actually using the G6Y relays, successor to the G5Y, same performance and footprint, but now also obsolete. The current G6K variant seems to have poor isolation even at 144 MHz, and of course an incompatible SM footprint. I am happy to make the original PCB design available for anyone who wants to change it - sorry its an Eagle design.

I think the NAIS relays are a pretty good alternative, in this position the mismatch is not so important as the isolation.

Interesting to see that the other frequency variants are being made.

73,

Hamish, OH2GAQ

On Wednesday, 1 July 2020 01:43:35 UTC+3, Duncan Clark wrote:

In message
<BYAPR01MB5016E19B2060DA2669A8C624E...@BYAPR01MB5016.prod.exchangelabs.c

On Wednesday, 1 July 2020 01:43:35 UTC+3, Duncan Clark wrote:

In message
<BYAPR01MB5016E19B2060DA2669A8C624E...@BYAPR01MB5016.prod.exchangelabs.c

[Marc OLANIE]

Hi Duncan & Hamish

 

Thanks a lot for the relay ref… the 50 Ohm modern version seems to be the ARE1012C90, mouser’s ref 769-ARE1012C90 (and it’s not to dear at 5 euros a piece without tax for a real RF relay)

Isolation is far from our wildest expectation  @ more than 80 dB

 

Hamish, you don’t have to worry concerning the file format. It is now possible to « import » an Eagle document and convert it with minor work, and it has never been a big problem to redesign a « almost one side » pcb using the Gerber file as a kind of stencil. So there are no real troubles here, only solutions 😊

If I had to redesign the pcb, I will send you all the Kicad files once the job done, to enrich your collection. You’re the only owner of you work.

I think it would be a nice move anyway, as a 70 cm version would need some helical filters instead a discrete bpf.

 

Thanks Duncan for the Coilcraft reference.

 

I’m still not able to simulate the 2m bpf with Elsie. with this typology, I obtain inductance values around 6nH , C shunt around 180 or 190 pf, and coupling series caps of 16 pf. Original values give me a rather small coupling and higher frequencies, even with a large variation of the inductor’s value.

 

Filter topology used : Nodal cap coupled bandpass, Butterworth, 15M bandwidth, 142M as central frequency, 3rd order.

 

My 20/31 MHz bpf is « almost » the same as yours. I choose to use fixed smd coilcraft inductors, and results with my N2PK VNA a close to the simulation.

 

Once again, a BIG thank you to Hamish for the fun we have building these transverters. It’s definitely a wonderful projetc

73’

Marc f6itu

 

Ps : hey, Duncan…reading at your answers, I should have guessed the « touch » of a PicaStar builder ;- ))

 

De : herme...@googlegroups.com <herme...@googlegroups.com> De la part de Hamish Kellock OH2GAQ
Envoyé : mercredi 1 juillet 2020 06:01
À : Hermes-Lite <herme...@googlegroups.com>
Objet : Re: Transverter Question

 

Hi Duncan and Marc,

 

I didn't realize this relay choice would be such a problem, originally designed the transverter using components that I had. I am actually using the G6Y relays, successor to the G5Y, same performance and footprint, but now also obsolete. The current G6K variant seems to have poor isolation even at 144 MHz, and of course an incompatible SM footprint. I am happy to make the original PCB design available for anyone who wants to change it - sorry its an Eagle design.

 

I think the NAIS relays are a pretty good alternative, in this position the mismatch is not so important as the isolation.

 

Interesting to see that the other frequency variants are being made.

73,

Hamish, OH2GAQ

On Wednesday, 1 July 2020 01:43:35 UTC+3, Duncan Clark wrote:

In message
<BYAPR01MB5016E19B20...@BYAPR01MB5016.prod.exchangelabs.c
om>, Marc OLANIE writes

In message
<BYAPR01MB5016E19B20...@BYAPR01MB5016.prod.exchangelabs.c
om>, Marc OLANIE writes

--

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/bbd424c5-f872-47d1-b98e-735af1e834b6o%40googlegroups.com.

[Duncan Clark]

In message
<BYAPR01MB5016C1F9BB...@BYAPR01MB5016.prod.exchangelabs.c

om>, Marc OLANIE writes

>I’m still not able to simulate the 2m bpf with Elsie. with this
>typology, I obtain inductance values around 6nH , C shunt around 180 or
>190 pf, and coupling series caps of 16 pf. Original values give me a
>rather small coupling and higher frequencies, even with a large
>variation of the inductor’s value.

I found the same when modelling in Elsie so I actually went with values
similar to G4DDK's Anglian 3L.

Series 3.3, 0.5, 0.5, 3.3pf

Parallel 4.7, 7.5, 4.7pf

[Marc OLANIE]

Hi

At last... modeling and simulation are possible with AADE filter Design (4.5). Topology "coupled resonator", Butt or Tcheb

The Q of this typology is lower than the one used by Elsie, but inductor's value is more "human" 😊

73'

Marc f6itu



-----Message d'origine-----

De : herme...@googlegroups.com <herme...@googlegroups.com> De la part de Duncan Clark
Envoyé : mercredi 1 juillet 2020 22:57
À : herme...@googlegroups.com
Objet : Re: Transverter Question

--
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/36zyCeM%24hP%24eFwua%40ntlworld.com.