TH2830 vs TH2832

[GnomeZA]

Well, that worked great. A nice consistent/smooth fan sound. It's quieter, more efficient, and the buzzing is gone.

My fan holes were 3mm.
I had to severely cut up the rubber grommets from Noctua to fit.
I wasn't down to drill out the holes in the instrument.

Alternatively I could have used some 3mm BHCS to mount them (original screws are far too short)
But I seeing Josh' picture I decided to give the grommets the opportunity.

---

If I take pure fan/air noise, the Noctua is louder than the original fan IMO.
The original fan had an annoying buzzing which made it far more annoying, but if you ignore that, the pure fan noise "part" of the sound is either same or slightly higher on the Noctua.
Overall I still think the upgrade is worth it, the fan moves a LOT more air and you don't have this annoying buzzing noise constantly going on.

Also Noctua fans are rated for a LOT of hours, it is unlikely the instrument will need a new fan in the next 20 years based on how much I use it.
Noctua is really up there in terms of their fan engineering (quite a few very nice videos about it on YouTube).

[KungFuJosh]

Well, that worked great. A nice consistent/smooth fan sound. It's quieter, more efficient, and the buzzing is gone.

My fan holes were 3mm.
I had to severely cut up the rubber grommets from Noctua to fit.
I wasn't down to drill out the holes in the instrument.

Alternatively I could have used some 3mm BHCS to mount them (original screws are far too short)
But I seeing Josh' picture I decided to give the grommets the opportunity.

I might have used the original screws inserted backwards to pull the metal mounts out of the chassis using an angled plyers. 😉

My stock fan was obviously older than yours, so for me it got quieter. It's comparatively quiet now, but same as you, the consistent sound is much less annoying than the buzz the stock fan had.

[GnomeZA]

(slight OT if people want to ignore)

Hello,

I too do not live in the US.

Noctua FANs are made in Austria. I find it usually best to order on their EBay shop.
Very fast and reliable service !

Best Regards,
rudi

Unfortunately getting things from America is just a lot easier than getting things from the EU.
For some reason everything is always more expensive and more work to get from the EU (EU stores are much less helpful with regards to export)

I don't get it because ordering from China and Japan (when things are actually made there), is cheaper unlike the EU situation.
So for example, ordering a Makita tool from Japan is much cheaper than the US.

But if you take Wera tools, they are more expensive ordering it from Germany than the US.
And not just shipping, the actual tool is cheaper from the US

So basically I don't even bothering checking EU prices anymore.

[KungFuJosh]

Here's a stock twenty eight thirty fw dated 12/2023. Hopefully we'll have a slightly different model's improved version if our friend @tv84 has some time. 😉

[GnomeZA]

Here's a stock twenty eight thirty fw dated 12/2023. Hopefully we'll have a slightly different model's improved version if our friend @tv84 has some time. 😉

Wow that is newer than even the one I got with my meter.
Have you tried it yet?
Which meter branding does it have?
How on earth did you get this? 

Any chance you could do some testing on that baby?

[KungFuJosh]

Wow that is newer than even the one I got with my meter.
Have you tried it yet?
Which meter branding does it have?

It's installed, and it has the ST branding on it.

How on earth did you get this? 

The ST people are AWESOME, that's how. That's why when anybody asks, I tell them to get the ST version if they can. The TH people will straight up ignore you until the end of time.

Any chance you could do some testing on that baby?

I plan to. 😉

[tv84]

Here's a stock twenty eight thirty fw dated 12/2023. Hopefully we'll have a slightly different model's improved version if our friend @tv84 has some time. 😉

Here, test and report.

Code: [Select]

Header: 414243446C0B070024951300
Signature: "ABCD"  (MAGIC OK)
Block1 Size:  00070B6C   [00000018-00070B83]
Block1 Checksums: 385C   [00070B84-000743DF]
Block2 Size:  00139524   [000743E0-001AD903]
Block2 Checksums: 9CA9   [001AD904-001B75AC]

Block1 - Checksum is OK!
Block2 - Checksum is OK!


[KungFuJosh]

Here, test and report.

You're awesome, thank you!!

Flashing went fine, and I'm happy to see I can do 100Ω DCR test now. I'll let it warm up, and then I'll do a full test with my test board.

Thanks!

[KungFuJosh]

Well, looks like things are better than they were for sure. This certainly appears to be the best firmware yet.

The bigger issues like 100Ω DCR and 100kΩ at 200kHz both look better.

The ~700mH inductor looks better than it was, but it still appears to be too far of a jump between those frequencies.

Results attached.

[GnomeZA]

The bigger issues like 100Ω DCR and 100kΩ at 200kHz both look better.

All the results in that document look well within tolerance.  Very promising!

The ~700mH inductor looks better than it was, but it still appears to be too far of a jump between those frequencies.

Ah you mean its inductance looks completely different between 1kHz and 10kHz?
With inductors and changing frequency like that, I would expect some inductors to behave in unexpected ways.
So I'm not completely sure whether to think this is a meter/calibration issue or more data needed.

Don't get me wrong, I'm not writing this off, just trying to think through it in my head.

All of this said, the results still look incredible!

[KungFuJosh]

Ah you mean its inductance looks completely different between 1kHz and 10kHz?
With inductors and changing frequency like that, I would expect some inductors to behave in unexpected ways.
So I'm not completely sure whether to think this is a meter/calibration issue or more data needed.

Don't get me wrong, I'm not writing this off, just trying to think through it in my head.

Yeah, I don't understand how they work well enough to speculate on that, but referring back to my previous test when I used the Shannon ST42 tweezers for reference:


The different for the ST42 in those two ranges was 37mH, but for the 2830 it was a difference of 150mH. Now the difference is only 100mH, which is still a bit much by comparison. Maybe somebody else has an excuse for it. 😉

All of this said, the results still look incredible!

Yes, I'm very happy with the improvements now. I'd say it's stable enough to be more trustworthy than any of their previous firmwares.

100KΩ at 100 and 200kHz is still very slightly out of spec, but not as bad as previously. I may test with a different fixture to see if that improves the accuracy in those ranges.

[Martin72]

Results attached.

Great work man, and this makes me laugh:


 

[GnomeZA]

I've made the switch on my meter to this firmware.
Thanks again @tv84 and also for the verification @KungFuJosh

[KungFuJosh]

Results attached.

Great work man, and this makes me laugh:


 

Me too 🤣

I'd love to know how though. Putting D on the screen isn't difficult, but I have no idea how to cycle through those ranges. Maybe they're based on the impedance range option? I dunno.

[Martin72]

This cannot be recorded in this way either; it is a calculated variable that only "works once", for example.
You had intended 1nF/1kHz for this, then you only get one value, namely the value for 1kHz.
In your next table, I would not include this as a range to be tested.

[KungFuJosh]

This cannot be recorded in this way either; it is a calculated variable that only "works once", for example.
You had intended 1nF/1kHz for this, then you only get one value, namely the value for 1kHz.
In your next table, I would not include this as a range to be tested.

I copied the note from the cal cert that was shared:

[Martin72]

Ok,

Then I have to say it the way you wrote it in your table.
It may also mean something other than the DF that comes to mind.
Edit:
Ah, this could be different 10nF caps...

[KungFuJosh]

Ok,

Then I have to say it the way you wrote it in your table.
It may also mean something other than the DF that comes to mind.
Edit:
Ah, this could be different 10nF caps...

I still have no idea how to test it. 😉

[Martin72]

It can only be that capacitors with different dissipation factors but the same capacitance have been used.
This is because the frequency and capacitance are the same for all ranges (10nF/1kHz).
But now find the right capacitors for it...
Or the different losses were created "artificially".
Tonghui should know.

[KungFuJosh]

Tonghui should know.

🤣🤣🤣 They should, but if they do, they'll never tell me. They're terrible.

I sent my results to my ST contact. If he knows, he might tell me.

[Martin72]

Please report then.

They should, but if they do, they'll never tell me

Same here with Hioki, East Tester...

[KungFuJosh]

Please report then.

They should, but if they do, they'll never tell me

Same here with Hioki, East Tester...

Even Hioki? Is it cause of it being an older model? That's disappointing. Hioki at least has firmware and stuff available on their website. Tonghui is douchebags about it to everybody. It's tedious to get anything from them, and certainly nothing has been available on their website.

[Martin72]

Even Hioki? Is it cause of it being an older model?

This is something I hadn't even posted here in the forum, I actually got a direct response from hioki earlier this year, probably because I contacted them with my company address this time.
Previously the distributors said there was nothing for my LCR because support was discontinued in 2015.
Now Hioki themselves had replied to me and written that they don't release anything to the public.
At least they are honest.
Tonghui and East Tester lie to you directly and say they have nothing.

[KungFuJosh]

Tonghui and East Tester lie to you directly and say they have nothing.

Yup. Tonghui told ST their newest fw was 2021. Then I sent ST a photo of Gnome's newer firmware version. After that is when TH gave ST the 2023 fw. I dunno if that's correlation or causation, but it's certainly interesting...and ridiculous.

[Martin72]

"Today" I got the new firmware for our ST2830 from Sourcetronic (the guy asked me if I knew about the new firmware through a certain "Josh"...  ).
I then asked again what exactly had been changed, the question was passed on and I may get an answer soon (which I will then post here).
This is interesting because I have to pay attention to the calibration.