Date   

Re: LO Board

Neil Smith G4DBN
 

I bought one of the original versions with the OLED display and found there was a PCB position for a reference input to replace the wobbly and drifty onboard XO.  It was usually within 10kHz or so, but the drift was terrible.  The software wasn't exactly reliable either, but it was cheap.  It felt like a student project rather than a proper commercial design.  Once I fitted the ref input, it was stable, but very noisy compared with an SV1AFN 4351 board.  I had to fit the SMA socket on the underside of the board and remove a link, also I killed the supply to the onboard XO.  I don't have itany more, I hadn't put it in a case and I dropped an SMA lead on the board while it was powered, and the magic smoke came out.  I gave the display board to someone who had a faulty one.

Maybe the software interface has improved now.  It had simple ramp generation but it felt like it needed a few weekends of serious debugging and a bit more UI development.  The ramps weren't easy enough to use to make it a pleasure, but it was sort-of OK as signal for checking out SHF receivers and for antenna alignment.  Lots of harmonics and quite loud at 10GHz.  I could even see harmonics at 24GHz on my specan.

Banggood does seem to have a decent reputation for ultra-cheap but useable machine tooling, they aren't like some of the fly-by-night ebay suppliers. I haven't bought anything electronic from them, but the suppliers I have used via Aliexpress and UTSource have all delivered non-fake devices, and for the higher-value kit, free express delivery in 4-5 days.  The power gasfets I bought for 5.7GHz are full spec and working fine, so I don't think it is fair to generalise about suppliers from the Far East.  There are good ones and those that are less good. The £11 FETs which I use in the DF9IC 23cm 250W PAs are stunning value, full props to UTSource for those.  Also had good devices from Teng Da, although their prices are not exactly low.

Buying from the Far East needs a bit of research and a bit of risk-acceptance.  It is a bit Wild West and there are charlatans aplenty, but also there are some businesses trading on their reputations. Not an investment for Widows and Orphans, as the saying goes.

I still prefer SV1AFN's kit though.

Neil G4DBN

On 12 May 2019, at 09:21, dave cossar via Groups.Io <davecossar=virginmedia.com@groups.io> wrote:

Hi Brian,

  The idea of having it all on the board appeals to me
But , if you look carefully  the new board does not have an input for
a voltage ref
reading the feedback  on Banggood I noticed that one guy had problems
getting an
accurate  output  and had n error of I KHz.


Re: LO Board

Colin G4EML
 

That should of course have read. ...oscillator to keep the cost down...


On 12 May 2019, at 09:39, Colin G4EML <colin@...> wrote:

Regarding the frequency error. Most of these low cost adf4351 boards have a poor stability reference oscillatory keep the cost down. If you need accurate frequencies you will have to either replace this with a better specified TCXO or feed the board from an external reference. 

Colin. 




On 12 May 2019, at 09:21, dave cossar via Groups.Io <davecossar@...> wrote:

Hi Brian,

 The idea of having it all on the board appeals to me

But , if you look carefully  the new board does not have an input for a voltage ref

reading the feedback  on Banggood I noticed that one guy had problems getting an

accurate  output  and had n error of I KHz.

 

This is one person only?

Dave GM3WIL

 

 


Re: LO Board

Colin G4EML
 

Regarding the frequency error. Most of these low cost adf4351 boards have a poor stability reference oscillatory keep the cost down. If you need accurate frequencies you will have to either replace this with a better specified TCXO or feed the board from an external reference. 

Colin. 




On 12 May 2019, at 09:21, dave cossar via Groups.Io <davecossar@...> wrote:

Hi Brian,

 The idea of having it all on the board appeals to me

But , if you look carefully  the new board does not have an input for a voltage ref

reading the feedback  on Banggood I noticed that one guy had problems getting an

accurate  output  and had n error of I KHz.

 

This is one person only?

Dave GM3WIL

 

 


LO Board

dave cossar
 

Hi Brian,

 The idea of having it all on the board appeals to me

But , if you look carefully  the new board does not have an input for a voltage ref

reading the feedback  on Banggood I noticed that one guy had problems getting an

accurate  output  and had n error of I KHz.

 

This is one person only?

Dave GM3WIL

 

 


Re: LO device?

Brian Flynn GM8BJF
 

Hi Dave,
That is a new board to me as it has the microcotroller chip on it along with an OLED display. 
Some of the earlier more basic boards  used pretty ordinary voltage regulators and were quite Phase noisy.
I see this one uses a Linear Technology chip which I take to be a low noise voltage regulator so the PN performance is probably OK.
The "black ADF4351 boards" use a generic LM1117 regulator which could be improved on
If it is an issue you can add LF decoupling to the power rails on the board.
 
The ADF4351 is quite useful chip and can provide a reasonably clean signal. A 4 GHz signal generator for £25 cant be too bad!
I made myself a simple sig gen using one of the "black boards" and Cypress USB interface which can be made to work
with the AD demo  software quite easily. I have had about a dozen of these sig gen boards from China and never had a problem with any of them.
They do what they say in the ads and are about a tenth of the price of the actual AD demo boards.
73s
Brian GM8BJF


Re: LMX2541 vs ADF4351

Richard <richard@...>
 



Thanks
a lot quicker and allows the ref doubler, so the phase comp is at 20 MHz 


Richard


On Sat, 2019-05-11 at 21:22 +0100, Andy G4JNT wrote:
Use my design tool
LMX2541_Prog.exe  contained in  http://g4jnt.com/LMX2541Support.zip

It can be used just to get the register values without having to use the RS232 interface.



On Sat, 11 May 2019 at 21:10, Richard <richard@...> wrote:
OK Andy
I just tried hard to get the ref divider down , but TI tool didn't want to play except with 2.5 and 2 MHz, I'll have another play and see if I can move the ref freq up
Richard


On Sat, 2019-05-11 at 21:03 +0100, Andy G4JNT wrote:
Close-in phase noise is purely a function of the reference.   Only when you get outside the loop bandwidth does the PLL chip come into the equation
In both you should be using Fcomp thd maximum you can - probably 10MHz, and a loop bandwidth in the 10 - 30kHz region


On Sat, 11 May 2019 at 20:56, Richard <richard@...> wrote:
Which in practice has the better close in phase noise
 Fvco =3321MHz, Fout =1660.5 Fref-10MHz  phase comp freq 2.5MHz
I'd forgotten I have quite a few that I got as samples, and one of
Andy's boards from a few moons ago
-- 
 Best wishes /73 
 Richard Bown
 
 Email : richard@...

 HTTP  :  http://www.g8jvm.com
 
 ######################################################################

 Ham Call: G8JVM . QRV: 50-432 MHz + Microwave 
 Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
 QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W, 
Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
 OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
 ######################################################################

-- 
 Best wishes /73 
 Richard Bown
 
 Email : richard@...

 HTTP  :  http://www.g8jvm.com
 
 ######################################################################

 Ham Call: G8JVM . QRV: 50-432 MHz + Microwave 
 Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
 QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W, 
Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
 OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
 ######################################################################


Re: LMX2541 vs ADF4351

Andy G4JNT
 

Use my design tool
LMX2541_Prog.exe  contained in  http://g4jnt.com/LMX2541Support.zip

It can be used just to get the register values without having to use the RS232 interface.



On Sat, 11 May 2019 at 21:10, Richard <richard@...> wrote:
OK Andy
I just tried hard to get the ref divider down , but TI tool didn't want to play except with 2.5 and 2 MHz, I'll have another play and see if I can move the ref freq up
Richard


On Sat, 2019-05-11 at 21:03 +0100, Andy G4JNT wrote:
Close-in phase noise is purely a function of the reference.   Only when you get outside the loop bandwidth does the PLL chip come into the equation
In both you should be using Fcomp thd maximum you can - probably 10MHz, and a loop bandwidth in the 10 - 30kHz region


On Sat, 11 May 2019 at 20:56, Richard <richard@...> wrote:
Which in practice has the better close in phase noise
 Fvco =3321MHz, Fout =1660.5 Fref-10MHz  phase comp freq 2.5MHz
I'd forgotten I have quite a few that I got as samples, and one of
Andy's boards from a few moons ago
-- 
 Best wishes /73 
 Richard Bown
 
 Email : richard@...

 HTTP  :  http://www.g8jvm.com
 
 ######################################################################

 Ham Call: G8JVM . QRV: 50-432 MHz + Microwave 
 Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
 QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W, 
Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
 OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
 ######################################################################


Re: LMX2541 vs ADF4351

Richard <richard@...>
 

OK Andy
I just tried hard to get the ref divider down , but TI tool didn't want to play except with 2.5 and 2 MHz, I'll have another play and see if I can move the ref freq up
Richard


On Sat, 2019-05-11 at 21:03 +0100, Andy G4JNT wrote:
Close-in phase noise is purely a function of the reference.   Only when you get outside the loop bandwidth does the PLL chip come into the equation
In both you should be using Fcomp thd maximum you can - probably 10MHz, and a loop bandwidth in the 10 - 30kHz region


On Sat, 11 May 2019 at 20:56, Richard <richard@...> wrote:
Which in practice has the better close in phase noise
 Fvco =3321MHz, Fout =1660.5 Fref-10MHz  phase comp freq 2.5MHz
I'd forgotten I have quite a few that I got as samples, and one of
Andy's boards from a few moons ago
-- 
 Best wishes /73 
 Richard Bown
 
 Email : richard@...

 HTTP  :  http://www.g8jvm.com
 
 ######################################################################

 Ham Call: G8JVM . QRV: 50-432 MHz + Microwave 
 Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
 QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W, 
Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
 OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
 ######################################################################


Re: LMX2541 vs ADF4351

Andy G4JNT
 

Close-in phase noise is purely a function of the reference.   Only when you get outside the loop bandwidth does the PLL chip come into the equation
In both you should be using Fcomp thd maximum you can - probably 10MHz, and a loop bandwidth in the 10 - 30kHz region


On Sat, 11 May 2019 at 20:56, Richard <richard@...> wrote:
Which in practice has the better close in phase noise
 Fvco =3321MHz, Fout =1660.5 Fref-10MHz  phase comp freq 2.5MHz
I'd forgotten I have quite a few that I got as samples, and one of
Andy's boards from a few moons ago
--
 Best wishes /73
 Richard Bown

 Email : richard@...

 HTTP  :  http://www.g8jvm.com

 ######################################################################

 Ham Call: G8JVM . QRV: 50-432 MHz + Microwave
 Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
 QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W,
 Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
 OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
 ######################################################################






LMX2541 vs ADF4351

Richard <richard@...>
 

Which in practice has the better close in phase noise
Fvco =3321MHz, Fout =1660.5 Fref-10MHz phase comp freq 2.5MHz
I'd forgotten I have quite a few that I got as samples, and one of
Andy's boards from a few moons ago
--
Best wishes /73
Richard Bown

Email : richard@g8jvm.com

HTTP : http://www.g8jvm.com

######################################################################

Ham Call: G8JVM . QRV: 50-432 MHz + Microwave
Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W,
Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
######################################################################


Re: LO device?

Andy G4JNT
 

RR 
The power of Google
I typed SV2AFN into Google to find the web site address to post, and it gave me the right page in spite of the typo
 


On Sat, 11 May 2019 at 16:30, F1CHF <F1CHF@...> wrote:
SV1AFN not SV2
here
regards,
 
nb :
indeed everything increases !
 
 
 
 
 
-------Message original-------
 
Date : 11/05/2019 15:13:08
Sujet : Re: [UKMicrowaves] LO device?
 
Yes, I've had two of that type (the black PCB) one of another Chinese low cost type (using a green PCB) and at least three of the decent properly made ADF4351 boards by SV2AFN

It was reported om in this group a few years ago - several people found that specific PCB, the black one, had a fake chip where the output divider couldn't go below 16.  Ie it was probably an ADF4350 device although labelled ADF4351.    Fortunately mine was going into a project that only needed to go down to 200MHz so teh fake chip was satisfactory

 The green PCB version was Ok

There are several versions of those cheap Chinese modules, some said to have better decoupling than others.

I'd never buy a Chinese one again, the SV2AFN ones are so much better made, properly decoupled and don't have fake chips.

As for the outpiut quality - it's mostly dependent on the quality of the reference you put in.  Raw 10MHz output from GPSDOs aren't always the best reference for synthesizers.



 


Réf. : Re: [UKMicrowaves] LO device?

F1CHF
 

SV1AFN not SV2
here
regards,
 
nb :
indeed everything increases !
 
 
 
 
 
-------Message original-------
 
Date : 11/05/2019 15:13:08
Sujet : Re: [UKMicrowaves] LO device?
 
Yes, I've had two of that type (the black PCB) one of another Chinese low cost type (using a green PCB) and at least three of the decent properly made ADF4351 boards by SV2AFN

It was reported om in this group a few years ago - several people found that specific PCB, the black one, had a fake chip where the output divider couldn't go below 16.  Ie it was probably an ADF4350 device although labelled ADF4351.    Fortunately mine was going into a project that only needed to go down to 200MHz so teh fake chip was satisfactory

 The green PCB version was Ok

There are several versions of those cheap Chinese modules, some said to have better decoupling than others.

I'd never buy a Chinese one again, the SV2AFN ones are so much better made, properly decoupled and don't have fake chips.

As for the outpiut quality - it's mostly dependent on the quality of the reference you put in.  Raw 10MHz output from GPSDOs aren't always the best reference for synthesizers.



Re: LO device?

Richard <richard@...>
 

the suppliers name should be a clue
Bl***y awful noisy goods
Richard

On Sat, 2019-05-11 at 14:20 +0100, Andy G4JNT wrote:
Ah, didn't look carefully enough.   That's a complete module with controller and OLED
Didn't see its a new version

Not sure I'd ever trust a low cost synth module from China ever again tho.

Not sure I'd trust anything from China these days



On Sat, 11 May 2019 at 14:13, Andy G4JNT via Groups.Io <andy.g4jnt=gmail.com@groups.io> wrote:
Yes, I've had two of that type (the black PCB) one of another Chinese low cost type (using a green PCB) and at least three of the decent properly made ADF4351 boards by SV2AFN

It was reported om in this group a few years ago - several people found that specific PCB, the black one, had a fake chip where the output divider couldn't go below 16.  Ie it was probably an ADF4350 device although labelled ADF4351.    Fortunately mine was going into a project that only needed to go down to 200MHz so teh fake chip was satisfactory

 The green PCB version was Ok

There are several versions of those cheap Chinese modules, some said to have better decoupling than others.

I'd never buy a Chinese one again, the SV2AFN ones are so much better made, properly decoupled and don't have fake chips.

As for the outpiut quality - it's mostly dependent on the quality of the reference you put in.  Raw 10MHz output from GPSDOs aren't always the best reference for synthesizers.



-- 
 Best wishes /73 
 Richard Bown
 
 Email : richard@...

 HTTP  :  http://www.g8jvm.com
 
 ######################################################################

 Ham Call: G8JVM . QRV: 50-432 MHz + Microwave 
 Maidenhead QRA: IO82SP38, LAT. 52 39.720' N LONG. 2 28.171 W
 QRV 6mtrs 200W, 4mtrs 150W, 2mtrs 300W, 70cms 200W, 
Microwave 1296MHz 140W, 2320MHz 100W & 10368MHz 5W
 OS: Linux Mint 19.1 x86_64 Tessa, on a Dell Inspiron N5030 laptop
 ######################################################################


Re: LO device?

Andy G4JNT
 

Ah, didn't look carefully enough.   That's a complete module with controller and OLED
Didn't see its a new version

Not sure I'd ever trust a low cost synth module from China ever again tho.

Not sure I'd trust anything from China these days



On Sat, 11 May 2019 at 14:13, Andy G4JNT via Groups.Io <andy.g4jnt=gmail.com@groups.io> wrote:
Yes, I've had two of that type (the black PCB) one of another Chinese low cost type (using a green PCB) and at least three of the decent properly made ADF4351 boards by SV2AFN

It was reported om in this group a few years ago - several people found that specific PCB, the black one, had a fake chip where the output divider couldn't go below 16.  Ie it was probably an ADF4350 device although labelled ADF4351.    Fortunately mine was going into a project that only needed to go down to 200MHz so teh fake chip was satisfactory

 The green PCB version was Ok

There are several versions of those cheap Chinese modules, some said to have better decoupling than others.

I'd never buy a Chinese one again, the SV2AFN ones are so much better made, properly decoupled and don't have fake chips.

As for the outpiut quality - it's mostly dependent on the quality of the reference you put in.  Raw 10MHz output from GPSDOs aren't always the best reference for synthesizers.




Re: LO device?

Andy G4JNT
 

Yes, I've had two of that type (the black PCB) one of another Chinese low cost type (using a green PCB) and at least three of the decent properly made ADF4351 boards by SV2AFN

It was reported om in this group a few years ago - several people found that specific PCB, the black one, had a fake chip where the output divider couldn't go below 16.  Ie it was probably an ADF4350 device although labelled ADF4351.    Fortunately mine was going into a project that only needed to go down to 200MHz so teh fake chip was satisfactory

 The green PCB version was Ok

There are several versions of those cheap Chinese modules, some said to have better decoupling than others.

I'd never buy a Chinese one again, the SV2AFN ones are so much better made, properly decoupled and don't have fake chips.

As for the outpiut quality - it's mostly dependent on the quality of the reference you put in.  Raw 10MHz output from GPSDOs aren't always the best reference for synthesizers.



2.4 GHz up converter

geoffrey pike
 

Just a quick sweep to show that the 2 filters take care of the LO at 1968 MHz
cheers
Geoff
GI0GDP




LO device?

dave cossar
 

HI Has anyone ever tried one of these from Banggood

 

 

https://uk.banggood.com/Geekcreit-Signal-Generator-Module-35M-4_4GHz-RF-Signal-Source-Frequency-Synthesizer-ADF4351-Development-Board-p-1416998.html?akmClientCountry=GB&gmcCountry=GB&currency=GBP&createTmp=1&utm_source=googleshopping&utm_medium=cpc_bgcs&utm_content=garman&utm_campaign=pla-gbg-ele-pc&gclid=EAIaIQobChMIhYrbnb2M4gIVaLftCh0vxAADEAQYASABEgL36PD_BwE&cur_warehouse=CN

 

 

Dave


Re: Using a M57762 ?

Nick Gregory G0HIK
 

Thanks Ian,
I'll use those ideas.
73s Nick G0HIK


GPS reference for the IC9700

Dave Brown
 

If you're interested in GPS referencing the IC9700, Glen has prototyped a
solution.

http://www.cortexrf.com.au/IC9700lock.html?fbclid=IwAR2QFKFOZ8U_4TpYjxpdYIjE
DlD8JIdhfpglEXoEBn7-4k7tSexgUCwyHCk

73
Dave, ZL3FJ


CQ CQ CQ 3cm 3cm Rain Scatter

G3UVR Denis IO83KH
 

Anybody like to try 3cm RS tonight? Worked G4ZTR on SSB just before Conditions good.

Replay here or meep KST de Denis G3UVR

14421 - 14440 of 64311