Products > Test Equipment

New EEVblog BM786 Multimeter

<< < (60/133) > >>

EEVblog:

--- Quote from: joeqsmith on January 23, 2021, 02:58:54 am ---
--- Quote from: EEVblog on January 22, 2021, 09:58:21 pm ---
--- Quote from: drussell on January 22, 2021, 02:26:35 pm ---Obviously the 786 and 789 will have different firmware since the 789 supports the LowZ AutoV, dBm, has T2, etc. I'm just curious if the one spec difference of the ACV frequency is an intentional firmware limitation or if it is related to actual filter components on the PCB, thats all.  My suspicion is that it is most likely a software limitation.
--- End quote ---
If they are smart then the firmware will be indentical, and only a single fuse bit determines what features it has.

--- End quote ---
What do you feel makes this the smart thing to do?  Makes it easier to maintain the code base?   Like the idea of promoting upgrades for beginners with a soldering iron?   Sell more meters?

--- End quote ---

Obviously from a code point of view it's highly preferable to have a single code base for all 3 meters in the series. So when you fix a bug or change/add a feature in one it gets fixed in all the others.
The code just reads a single flag like "if 786 or 789 then do this" etc.
You'd have to be crazy to want to maintain and compile three entirely different code bases for effectively the same meter. The model bit just gets set in the programmer software in a non-volatile register somewhere and the code reads that as required and either ignores or does that function that models has.

Kleinstein:
One can have a common code base with a check in the actual meter, or one can do the different versions at compile time, e.g. with a constant to change between versions - this would need less momory in the meter (or at least be lowest range one).

I would not expect the AC frequency limit to be a software point. AFAIK they use hardware RMS chips, so there is not much to adjust in software. The difference would be more a thing of the used amplifer and RMS converter grade, maybe filter caps. Even for the numerical RMS way that some chip sets more support, there is usually no provision for extra filtering, other than maybe choose the sampling rate. It could be a nice feature to have a choosable BW for RMS. Higher BW also means more noise / background.  After all his is a DMM and not a spectrum analyser - so don't expect too much.

HKJ:
If the controller is programmed in-circuit, there is no problem with separate firmware builds for the different models, but they could still use the same code base.
If the controller is programmed before mounting it on the PCB, it would be easier to use a common code for all models and then use a bit in the calibration memory or a link on the circuit board to select the actual model.

The 100kHz bandwidth requires compensation capacitors for the input stage, the 7kHz probably not.

2N3055:
Next to the LCD controller, there is an empty footprint for an IC and its components..
By the looks of it, it might be AD8436 True RMS converter chip. Some of the multimeter chips have built in low bandwidth RMS converters, for low cost options, with option to add external higher performance RMS converter for better specced meters.
Here we might have the same situation..

joeqsmith:

--- Quote from: EEVblog on January 23, 2021, 05:26:13 am ---
--- Quote from: joeqsmith on January 23, 2021, 02:58:54 am ---
--- Quote from: EEVblog on January 22, 2021, 09:58:21 pm ---
--- Quote from: drussell on January 22, 2021, 02:26:35 pm ---Obviously the 786 and 789 will have different firmware since the 789 supports the LowZ AutoV, dBm, has T2, etc. I'm just curious if the one spec difference of the ACV frequency is an intentional firmware limitation or if it is related to actual filter components on the PCB, thats all.  My suspicion is that it is most likely a software limitation.
--- End quote ---
If they are smart then the firmware will be indentical, and only a single fuse bit determines what features it has.

--- End quote ---
What do you feel makes this the smart thing to do?  Makes it easier to maintain the code base?   Like the idea of promoting upgrades for beginners with a soldering iron?   Sell more meters?

--- End quote ---

Obviously from a code point of view it's highly preferable to have a single code base for all 3 meters in the series. So when you fix a bug or change/add a feature in one it gets fixed in all the others.
The code just reads a single flag like "if 786 or 789 then do this" etc.
You'd have to be crazy to want to maintain and compile three entirely different code bases for effectively the same meter. The model bit just gets set in the programmer software in a non-volatile register somewhere and the code reads that as required and either ignores or does that function that models has.

--- End quote ---
I agree about using a common code base and conditional builds.  Of course, if it's the identical firmware there's no need for that. 

I never asked them why they spun up different firmware.  To me, it doesn't matter as long as they are able to maintain it.  Based on how well they kept up with me, I don't see that as a problem.   UEI had one meter to maintain and couldn't pull it off.   I think there's a difference in the skill level of the developers which may be a much bigger factor. 


--- Quote from: 2N3055 on January 23, 2021, 03:14:33 pm ---Next to the LCD controller, there is an empty footprint for an IC and its components..
By the looks of it, it might be AD8436 True RMS converter chip. Some of the multimeter chips have built in low bandwidth RMS converters, for low cost options, with option to add external higher performance RMS converter for better specced meters.
Here we might have the same situation..

--- End quote ---

I didn't spend any time trying to sort out the differences beyond changing the IC.   While apart, I noticed the additional shield and trimmers.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod