NanoVNA V2 / Forum

Note: this page is a mirror of https://groups.io/g/NanoVNAV2.
Click here to join and see most recent posts.

Updated modified firmware for the 4" V2.74 from blackmagic (V2+)


kd4pbs 2020/10/30 11:58

Good afternoon, NanoVNA friends!

I modified my SAA2N (compatible firmware for the 4" V2.74 from blackmagic) to the V2+ spec by performing the capacitor/resistor mod to upgrade from V2 to V2+.  Now I'm in need of firmware which is compatible with this modification to take advantage of the faster sweep rate.  I could of course roll my own, but at first glance, I'd have to spend a bunch of time with my current distribution to add the needed libraries, etc. for installation of the compiler for the arm platform.  So, I'm trying to take the lazy route by finding out if someone has already built a ready-to-go firmware for such a combination.

Regards,
Matt

OwO 2020/10/31 12:09

There is a 4" v2plus (not to be confused with v2plus4) firmware build:
https://nanorfe.com/downloads/20201013/nanovna-v2-20201013-v2plus-st7796.bin

ea4bfk 2020/10/31 01:34

HI all

Where I can find the details of the modification "capacitor/resistor" to upgrade my VNA 2_2?.Mine is a version 2.2 with SMA connectors that I change to 4" display. Firmware V2 Plus from Oct 13, 2020 works, but I don't now if i need to make the modification.

Regards
Alex

EA4BFK

hwalker 2020/10/31 06:04

On Fri, Oct 30, 2020 at 09:07 PM, kd4pbs wrote:

I modified my SAA2N (compatible firmware for the 4" V2.74 from blackmagic) to the V2+ spec by performing the capacitor/resistor mod to upgrade from V2 to V2+.  Now I'm in need of firmware which is compatible with this modification to take advantage of the faster sweep rate.  I could of course roll my own, but at first glance, I'd have to spend a bunch of time with my current distribution to add the needed libraries, etc. for installation of the compiler for the arm platform.  So, I'm trying to take the lazy route by finding out if someone has already built a ready-to-go firmware for such a combination.

=====================================

Matt,
In addition to the hardware modification, my understanding is that you will also need to also change the bootloader before the new firmware will work properly.  Cho45 has provided detailed instructions for the entire modification at:

https://lowreal.net/2020/10/28/1

Use Goggle translate to translate from Japanese to English.

- Herb

hwalker 2020/10/31 06:29

On Sat, Oct 31, 2020 at 06:19 AM, ea4bfk wrote:

>
> Where I can find the details of the modification "capacitor/resistor" to
> upgrade my VNA 2_2?.Mine is a version 2.2 with SMA connectors that I
> change to 4" display. Firmware V2 Plus from Oct 13, 2020 works, but I
> don't now if i need to make the modification.

Alex,
The firmware you loaded will work, but without the hardware modification and bootloader you will should be seeing an increase in the noise and no increase in speed.

The following link will give you full details regarding both the required hardware and bootloader modifications.

https://lowreal.net/2020/10/28/1

- Herb

The previous link I provided at

pe2kxh 2020/10/31 23:11

hello

Is there also a english version of the modification... my japanese is not
that good :-)

regards, jac pe1kxh

On Sat, Oct 31, 2020 at 2:29 PM hwalker <herbwalker2476@gmail.com> wrote:

Mike Runyan 2020/10/31 15:34

https://github.com/eried/Research/tree/master/NanoVNA/v2/upgrade_v2.2_to_plus

Neal Pollack 2020/10/31 15:56

The very first link that you see in the center of that page, the only words
in english,
that link takes you to the english project page.

Cheers

Neal

On Sat, Oct 31, 2020 at 3:11 PM pe2kxh <jemsmeets@gmail.com> wrote:

Siegfried Jackstien 2020/11/01 01:08

You can use google translate... That worked for me

Greetz sigi dg9bfc



Am 31.10.2020 23:34 schrieb "Mike Runyan via groups.io"
<therunyans=mac.com@groups.io>:

>
<https://github.com/eried/Research/tree/master/NanoVNA/v2/upgrade_v2.2_to_plus>



_._,_._,_

* * *

kd4pbs 2020/11/02 13:35

Thanks, OwO.  And thanks for your work on this.

I installed the firmware you linked and tested the results.  Unfortunately, the lowered SNR / dynamic range is unacceptable for me for this modification.  Perhaps the SAA2N board is not a very good copy of the SAA2?  I'm not sure.  Regardless, there is a noticeable degradation in performance of the unit in every aspect other than the increase in sweep speed.

Oh well, it was worth a try.

Return loss, smith chart, and isolation, S11 and S21 into cal load, 2.2Plus mode:

Return loss, smith chart, and isolation, S11 and S21 into cal load, 2.2 mode:

Smith Chart, S11 into calibration load, 2.2Plus mode:

Smith Chart, S11 into calibration load, 2.2 mode:

Return loss, S11 into cal load, 2.2Plus mode:

Return loss, S11 into cal load, 2.2 mode:

Return loss and through, S11+S21, 2.2Plus mode:

Return loss and through, S11+S21, 2.2 mode:

Regards,

Matt

hwalker 2020/11/02 14:39

On Mon, Nov 2, 2020 at 01:35 PM, kd4pbs wrote:

>
>
>
> Thanks, OwO.  And thanks for your work on this.
>
>
>
> I installed the firmware you linked and tested the results. 
> Unfortunately, the lowered SNR / dynamic range is unacceptable for me for
> this modification.  Perhaps the SAA2N board is not a very good copy of the
> SAA2?  I'm not sure.  Regardless, there is a noticeable degradation in
> performance of the unit in every aspect other than the increase in sweep
> speed.
>
>
>
> Oh well, it was worth a try.
>
>
>
>

Matt,
See Gabriel's response in message #169.

" V2 plus has many more hardware changes than the published "3 capacitors/resistors" upgrade, which doesn't completely upgrade your device to v2plus but allows you to get the sweep speedup. V2Plus firmware is still open source and you can read the code to see how the sweep speed is increased. The noise improvements (S21 noise floor) is due to other hardware changes though."

So installing the firmware and "3 capacitors/resistors" upgrade will get you the sweep speedup but will result in degraded performance without the other hardware changes. Best to either be satisfied with the performance of your current V2 device, or purchase a V2Plus if you need the additional noise improvements.

- Herb

kd4pbs 2020/11/02 14:47

I did stumble across that post after I posted this.  Makes sense.

At least this info is there now for anyone else wishing to perform this modification.

Regards,

Matt

OneOfEleven 2020/11/02 18:45

That's quite a difference in the noise floor/dynamic range. Think I'll not modify the V2's I have to V2+ in that case, unless the other differences can be done too (assuming they are only component changes) ?

Siegfried Jackstien 2020/11/03 04:29

You made the fw upgrade but not the hardware modding?? No wonder it did have
lower performance...

You have to do the hardware mod... And then do the fw change

And after a calibration a matched load should show a perfect dot in centre

Uncalibrated... And no parts change it shows not what you would expect



Greetz sigi dg9bfc

_._,_._,_

* * *

OwO 2020/11/03 14:31

It looks like that if you update to v2plus firmware without the hardware
changes
(https://github.com/eried/Research/tree/master/NanoVNA/v2/upgrade_v2.2_to_plus),
after the hardware changes the noise floor should be identical to before.

kd4pbs 2020/11/04 10:13

kd4pbs 2020/11/04 10:34

And, just to explain and make sure we are all on the same page here, the procedure was:

* Modify the SAA2N by changing the two capacitors and one resistor
* ohm out all componentry and traces to ensure this was done properly
* Realize that I didn't know where to find the firmware to take advantage of this mod
* Come here and explain the issue of not knowing where to find the proper firmware
* Install the compatible 4" 2+ firmware which OwO linked (thanks!)
* Calibrate (SOLT)
* Take measurements and snap a picture of them
* Install the 4" v.2 firmware I had before (the "V2.74" from blackmagic 20201013 build from https://nanorfe.com/nanovna-versions.html )
* Calibrate (SOLT)
* Take measurements and snap a picture of them
* post findings here.

If there is a better methodology someone might have in mind, I'm all ears.  Just understand that using, driving, and interpreting results from a VNA is part of my day job that I've had for the past 32 years...

Siegfried Jackstien 2020/11/04 19:29

.... you have to decide what you want

if you want a plus ... change hardware and use plus firmware

if you want to use the non plus firmware then do not that hardware change

you can not mix them!!

so maing the hardware mod ... then use plus fw ... and then upload
"geniune (nonplus) fw is not correct

if you want to compare then

first do the sweep with genuine board and fw

then do hardware changes and upload plus fw

and then do the next sweep

BUT ... the two firmware have also different filter width and averaging
so to compare (and have a fair result) you have to switch average off in
the v2 but use average x2 in the plus (gabrielle just told me that an
hour ago in discord discussion/chat)

and ... to really have a v2plus you also need that special bootloader
cause the bl sets some things in the fw that you would not have with the
"genuine" bootloader

greetz sigi dg9bfc

Am 04.11.2020 um 18:34 schrieb kd4pbs:

kd4pbs 2020/11/04 12:43

Hi, Siegfried,
Indeed, I get where you're coming from on that and agree - I should have taken some pre-mod measurements as well.  I'll "de-mod" the VNA and post those results too.  I think that turning on averaging on one version and turning it off on another would be an unfair comparison, and also pointless in that it doubles the time it takes for a sweep back to V2 speed; I think the point of the plus version is increasing the sweep speed while not degrading any other performance whatsoever.
I did not realize that the bootloader has to change - I'll update that and compare as well.

Regardless, it is apparent that changing between the V2 firmware and the V2+ firmware with the hardware modification brings me much worse SNR and dynamic range on this modified SAA2N, which is exactly the point of my first post.  I'll certainly find out if this is rectified with a simple change of the bootloader.

And let me also clarify; *this is NOT a V2+ device*.  I'm sure the actual V2+ performs great... perhaps someone with a genuine V2+ would be so kind as to make some similar measurements and post those results.

Siegfried Jackstien 2020/11/05 00:08

i can only repeat what gabriel told me ... use x2 avg on the v2plus to
compare it with a genuine v2 ... its not only the scan speed that is
changed but also other things ...

if after the v2 upgrade and using the avg x2 shows lower noise ... then
it was worth it even if then overall scan speed is the same

but change bootloader also is tricky

Am 04.11.2020 um 20:43 schrieb kd4pbs:
>
> Hi, Siegfried,
> Indeed, I get where you're coming from on that and agree - I should
> have taken some pre-mod measurements as well.  I'll "de-mod" the VNA
> and post those results too.  I think that turning on averaging on one
> version and turning it off on another would be an unfair comparison,
> and also pointless in that it doubles the time it takes for a sweep
> back to V2 speed; I think the point of the plus version is increasing
> the sweep speed while not degrading any other performance whatsoever.
> I did not realize that the bootloader has to change - I'll update that
> and compare as well.
>
> Regardless, it is apparent that changing between the V2 firmware and
> the V2+ firmware with the hardware modification brings me much worse
> SNR and dynamic range on this modified SAA2N, which is exactly the
> point of my first post.  I'll certainly find out if this is rectified
> with a simple change of the bootloader.
>
repeat ... change bl IS NOT THAT SIMPLE


> And let me also clarify; *this is NOT a V2+ device*.
>
it is .... "almost" ... but you made only half of the modding and you
compared apples with oranges hi hi

> I'm sure the actual V2+ performs great... perhaps someone with a
> genuine V2+ would be so kind as to make some similar measurements and
> post those results.
>
in a few days i will get a v2plus4 ... then modify one of my v2 to a
v2plus and then we have 3 different evolution stages to compare

a genuine v2, a v2plus ... and a v2plus4

i will post my results ... but will do that when that new v2plus4 is here

greetz sigi dg9bfc

kd4pbs 2020/11/05 08:27

Very good.  Comparisons are key here, apples to apples.And for fun, I'll certainly do the 2x average on the "2ModPlus" as well.

Why is changing the bootloader tricky?  I thought it was a simple matter of using the proper programs, programmer, and bootloader file?  For sure it might be tricky locating the bootloader file I require, but I do have an ST Link V2 programmer.

Perhaps later today I can get to this.

Thanks, Sigi.

Regards & 73,

Matt

ea4bfk 2020/11/05 11:25

Hi all

After receiving the information for the modifications needed from Sigi, I made several measurements with original firmware for 4", that was posted in the group and then with the new firmware, without and with the modifications.

In the attached pdf you will see my findings.

All measurements has been done with VNA QT, as nanSaver doesn't work with the 2Plus firmware.
Regarding the impact of the changes, I've seen and increase on noise in S11 measurement with the new firmware for 2Plus, also after making the hardware modification.
Also there is the peak around 2.685 GHz on Rx port (Port1) in Open and short situation, that I guess can be related potentially with my enclosure.
Please go trough the document and please provide me your comments.

Thanks to all

Alex
EA4BFK

Mike Runyan 2020/11/05 13:09

@EA4BFK,
The latest nanoSaver firmware works.  Or turn averaging off and save to slot 0 before hooking up nanoSaver.

Siegfried Jackstien 2020/11/05 22:26

that is a good hint to turn avg off before connecting to saver ... a
fault that should be reported on github (and i do it in next minute)

greetz sigi dg9bfc

Am 05.11.2020 um 21:09 schrieb Mike Runyan via groups.io:

ea4bfk 2020/11/05 15:46

Thanks Mike

Tomorrow I’ll try.

Alex
EA4BFK

Stephen Laurence 2020/11/06 05:15

Dear Matt,

A fabulous set of pictures of your V2N in various states. I must keep them for reference.

I have four V2 devices in various guises, including a V2N. NONE of them have as good a sweep as your screenshot of the unmodified device. I would say that, your specimen is an excellent one and certainly not a clone. The V2N devices, I think are not popular enough to be worth cloning.

My V2N is about the same in and out of the case. My first v2 (red display board and no case) is being used as a test board to reduce noise. Currently it has got loads of hard ferrite slabs, 1-2mm thick, all over both sides of the board, with some modest improvement but still nowhere near as good as yours.

Steve L.

OneOfEleven 2020/11/06 06:51

I've just upgraded my NanoVNA V2 to V2Plus and all is good.

New battery has the one that came with it set on fire (lots of smelly smoke and flames) :(
The resistor and two capacitors were changed.
Replaced boot loader with the one meant for V2Plus.
Uploaded V2Plus firmware (Oct 13 2020).
It now does 203 points/sec (100/points/sec before) - see App image.
The noise floor/dynamic range still seems good (see photo - 15dB per division on both S11 mag and S21 mag).

OneOfEleven 2020/11/06 06:51

[Edited Message Follows]

I've just upgraded my NanoVNA V2 to V2Plus and all is good.

New battery has the one that came with it set on fire (lots of smelly smoke and flames) :(
The resistor and two capacitors were changed.
Replaced boot loader with the one meant for V2Plus.
Uploaded V2Plus firmware (Oct 13 2020).
It now does 203 points/sec (100/points/sec before) - see App image.
The noise floor/dynamic range still seems good (see photo - 15dB per division on both S11 mag and S21 mag).

OneOfEleven 2020/11/06 06:53

missing images ..

Stephen Laurence 2020/11/06 08:21

At my end, I got the pictures three times. Sometimes this forum does funny things.

It looks promising, especially as the points per sec are doubled. However did you do the 50k to 900meg scan before any mods?

The noise floors on my unmodified early V2 up to 900meg is about the same as yours, but on Matt’s V2N a tiny bit better than yours. Have you done a sweep up to 3ghz, or 3.8ghz yet, as the higher frequencies are where the noise floor gets worse and an improvement especially at the top end would be worth doing the mods for.

I am a bit alarmed about the battery fire though. The flat batteries are in every phone, Ipad and tablet and fortunately they seem to rarely give trouble. I did buy a nonworking Iphone 3 where the battery had swollen. I hoped to get it going (it had a big memory) but the battery expansion had stressed the processor board and it was unsalvageable. Replacement Iphone 4 and 5 batteries are not too expensive and I was thinking they would fit in the V2 nsnos.

Steve L

OneOfEleven 2020/11/06 09:34

Here's a quick calibration for 50kHz to 3GHz, I forgot to set it to 201 points but no matter. This is with 50R SMA loads on both ports, and the scale is 15dB per div ..

OneOfEleven 2020/11/06 09:35

Maybe it's a little different due to it not having a metal case ? .. not sure.

OwO 2020/11/07 01:44

The battery fire doesn't surprise me at all, in fact I am more surprised
there aren't more reports of NanoVNA battery fires. Even with my
supervision in QC and two layers of protection (protection IC +
polyfuse) in later V2Plus versions I am still not confident enough to
allow an included flat li-ion battery. Only starting from plus4 with an
18650 cell did I sign off on allowing an included battery. 18650 cells
have a rigid outer shell, unlike a pouch cell which will explode if
poked, and also have more inbuilt protection (mechanical thermal cutoff)
than pouch cells which have none and completely rely on electrical
protection.

Earlier NanoVNAs (V1) have the explosion designed in from the start, it
charges the battery at over 2C which is way out of the specification of
the included batteries.

Peter KA6Z 2020/11/07 14:29

Hmmm, should we think about replacing the flat batteries in our pre-Plus4 VNAs with 18650s in holders epoxied to the back of our units?

Siegfried Jackstien 2020/11/07 22:38

for better safety i would say yes

i had an idea for an adaptor board ... holds a 4 inch and also has a
cutout for 2 round cells

just a rough sketch to give the idea ... add some port extenders (male
female) or sma to n adaptors

new battery switch and new pushbuttons??

at least you have the 4 inch well centered and get rid of that screen
cracking flat pack ... maybe i have drawn the buttons a bit to far out
but as noted it is just a sketch ...any v2 owner can now upgrade to 4
inch and replace that flatpack with 1000 mah with two round celld with
5000 to 6600mah ... then the carge current also is safer if you push it
in 5 to 6 ah cell and not in a 1 ah cell ;-)

dg9bfc sigi

Am 07.11.2020 um 22:29 schrieb Peter KA6Z via groups.io:

Dragan Milivojevic 2020/11/08 00:10

If you charge your nano via the USB port there is not much to worry about.
Most USB2 ports are limited to 500mA.

On Sat, 7 Nov 2020 at 23:29, Peter KA6Z via groups.io <p.knazko=
yahoo.com@groups.io> wrote:

Peter KA6Z 2020/11/07 23:25

Hi Dragan, so I wonder how else the fire could happen, since usually the only way available to charge these batteries is USB? I suppose some of these flat batteries are just of poor quality and if we could reliably source ones we know are of high quality then the problem could be largely avoided, without resorting to bolting on 18650s instead.
On Saturday, November 7, 2020, 03:10:21 PM PST, Dragan Milivojevic <d.milivojevic@gmail.com> wrote:

If you charge your nano via the USB port there is not much to worry about.Most USB2 ports are limited to 500mA.

On Sat, 7 Nov 2020 at 23:29, Peter KA6Z via groups.io <p.knazko=yahoo.com@groups.io> wrote:

Hmmm, should we think about replacing the flat batteries in our pre-Plus4 VNAs with 18650s in holders epoxied to the back of our units?

Dragan Milivojevic 2020/11/08 01:11

Even good batteries sometimes burst into flames,
remember the exploding Samsung phones a few years back?
Only readily available source of good batteries, that I know,
are the ones from mobile phones.


On Sun, 8 Nov 2020 at 00:25, Peter KA6Z via groups.io <p.knazko=
yahoo.com@groups.io> wrote:

Jim Lux 2020/11/07 20:28

On 11/7/20 3:25 PM, Peter KA6Z via groups.io wrote:
> Hi Dragan, so I wonder how else the fire could happen, since usually the
> only way available to charge these batteries is USB? I suppose some of
> these flat batteries are just of poor quality and if we could reliably
> source ones we know are of high quality then the problem could be
> largely avoided, without resorting to bolting on 18650s instead.

It's not so much a quality, as much as it's a handling thing. - mind you
a "low quality" retailer/shipper is really about handling.

Pouch batteries are inherently sensitive to mistreatment (where
mistreatment is actually pretty mild) - Since the Robosimian fire in the
lab at JPL, we have to have all kinds of special permission and
precautions to even order them. They are VERY sensitive to being
squished or bent in handling.

https://www.wired.com/2016/10/samsung-isnt-one-lithium-ion-problems-just-ask-nasa/

That's an ok constraint in manufacturing phones or computers where
you've got a well defined assembly line that can handle all the
batteries exactly the same.

Less so if the battery is shipped in a padded envelope with bubble wrap.




kd4pbs 2020/11/09 08:35

Just checking in with the group - a huge last-minute work project has whittled away some of my time to work on the NanoVNA project, but hopefully I will be able to get to it before our next big work project next week.

I did diddle with it a little last week when I de-modded and then re-modded the NanoVNA, and my findings were that there wasn't much difference in the SNR and dynamic range in the de-modded (original) configuration and the modded (resistor and capacitor changes) mode with the 10132020 V2 (non-plus) firmware.  I re-modded it and kept it out of it's metal case to take a few measurements, and as some others have noticed, there indeed is noticeable increase in noise and decrease in dynamic range with it inside a metal case - mainly in the >1GHz range.

73,

Matt

OneOfEleven 2020/11/10 03:32

On Sat, Nov 7, 2020 at 11:25 PM, Peter KA6Z wrote:

>
>
> Hi Dragan, so I wonder how else the fire could happen, since usually the
> only way available to charge these batteries is USB? I suppose some of
> these flat batteries are just of poor quality and if we could reliably
> source ones we know are of high quality then the problem could be largely
> avoided, without resorting to bolting on 18650s instead.
>

The VNA that the battery turned into a firework, the flame was like a small jet out the one side of the battery (so didn't harm the PCB), short lived but fast and furious, masses of smoke.
The VNA was not cased, it came like that, so has no physical protection from the outside world, and so is very easy to nick or pin prick the battery in all sorts of ways. The moment it happened I took it straight outside and left it there in the garden (middle of the path) till the next day where it could no further harm.

The other two V2's I have are in their metal boxes which I like very much, it makes them very much safer if ever their batteries did the same, even if the metal cases do reduce the performance slightly I'd rather them now be safe to leave about on their own than to set something else nearby alight. At least the batteries are protected by a metal barrier.

The smell they create is terrible I know that much !

If any of your VNA's are also not cased in metal, be extremely careful !

Hector Pascal 2020/11/18 17:59

On Sat, Nov 7, 2020 at 08:28 PM, Jim Lux wrote:

>
> https://www.wired.com/2016/10/samsung-isnt-one-lithium-ion-problems-just-ask-nasa/
>

Jim Lux 2020/11/18 20:19

On 11/18/20 5:59 PM, Hector Pascal wrote:
> On Sat, Nov 7, 2020 at 08:28 PM, Jim Lux wrote:
>
> https://www.wired.com/2016/10/samsung-isnt-one-lithium-ion-problems-just-ask-nasa/
> <https://www.wired.com/2016/10/samsung-isnt-one-lithium-ion-problems-just-ask-nasa/>
>
> Very scary video, but Robosimian DID have 96 Li-ion batteries
> installed!   Luckily the techs had gone for lunch.
> At the end of the wired.com piece by Sarah Scoles, is a
> light-hearted comment that nevertheless, we should all take seriously:
> "Because you know what they say: With great power comes great
> responsibility."
> H-P
>

even better, it was a summer hire intern who pushed it outside.

Siegfried Jackstien 2020/11/21 20:45

REPEAT ... ALSO BOOTLOADER NEEDS TO BE CHANGED AND THAT IS MISSING IN YOUR LIST!!!

To reply to this topic, join https://groups.io/g/NanoVNAV2

View this thread on groups.io