Blackmagic Firmware

[Mandy Geise]

BlackmagicDesign released a firmware update package for the Blackmagic eGPU/Pro. What interests me the most is that it's for macOS. I downloaded and installed the package but without the applicable eGPU I couldn't see the process. I found the folder that contains new 1.2 firmware bin files. It would be great to figure out where the firmware update tool is and whether we could use it for other TB3 enclosures. I have heard from a TB3 vendor that Apple has this firmware update tool for a while.

And requires restarting the system to apply the update. I don't see a firmware updater itself in the package which likely means that macOS itself might have a firmware update mechanism (and it reads from the directory as shown). I'll look a bit more.

Can't say much right now as I think it would be good to get some insight/see how the installer officially works. The plist has information such as vendor/device name and model numbers. So maybe we could substitute/create another entry for different enclosures and try.

Edit: yep, that's all this "installer" does - places the files into /Library/ThunderboltAccessoryFirmwareUpdates folder and then tells you to restart. There is no check for Blackmagic eGPU, etc. Plain old file copy. Probably a sufficient indicator that the firmware updater actually lies in macOS itself. The minimum requirement for the firmware update to work is 19A583 which is macOS 10.15.

@mac_editor Thank you for taking a look into the installer package. It would be great if a Blackmagic eGPU found this topic and make observation during the firmware update process and let us know. As you mentioned making changes to the plist file to match an enclosure bin firmware file is possibility but we don't know what else macOS is verifying during the flashing process.

Edit: Was also able to reproduce notification to update Mantiz Venus. It seems the tool is called ThunderboltAccessoryFirmwareUpdater which uses MobileAccessoryUpdater:

For kicks I tried flashing Venus firmware to the Sonnet box to see what happens. I noticed something important - when rebooting, I saw the progress bar for a second (meaning there was an attempt to flash firmware). Sadly my PD is still 15W so it failed. However, I gleaned much more insight, and now know the EFI app responsible for executing firmware flashes as well as its arguments:

@itsage I tried again with the Aorus box. This time, I observed something a little different: when rebooting, the progress bar came up, filled up all the way for a moment (probably successful?) and then rebooted. Result from NVRAM:

As of 2022 there are many badly performing clones on the market. V2/3GHz NanoVNA uses parts like ADF4350 and AD8342 which are costly and clones have been cutting costs by using salvaged or reject parts.

See official store and look for V2 Plus4/V2 Plus4 Pro versions only to avoid getting a bad clone. We have stopped selling V2.2 versions since October 2020, so all V2 hardware that are not Plus or Plus4 are not made by us and we can not guarantee performance.

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

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) ?

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.

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

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

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.

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.

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 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 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?

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

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 !

3a8082e126