SkyRC — IFA 2014 — MC3000 charger-analyzer

[mrQQ]

So you are saying you're going to teach all the end users to use USB debug viewer or...?

I mean, this is a bug. It's a huge problem. And it's not related to USB in any way, because it's a problem with logging: when time displays 10:00:00 in UI, it writes 09:36:00 in the file. That's the issue, not some USB yadda yadda.

 

[Albert_]

(stuff deleted)

I think it's a USB vulnerability.

 

[mrQQ]

You do understand, than PCLS shows 10:00:00, which is correct time? And that same second 09:36:00 is being written to file, which is incorrect time. So if the monitor app DOES know correct time, then how is this USB fault?

 

[Albert_]

(stuff deleted)

It's not USB's fault. USB not primarily designed for data sensitive applications. One of it's primary functions is the capability of using with with many different type of plug n play devices. Rapidly inserting and ejecting cable. There are many features in it's architecture not found in a typical RS-232 data cable and devices, which is designed more for stationary applications and devices, before there was any plug n play.

 

[kreisl]

I mean, this is a bug. It's a huge problem.

I don't have such a bug, the maker team doesn't have the bug, we don't have that problem. And we cannot reproduce your bug/problem, so it remains "UNCONFIRMED".

So your device counts time correctly and measures mAh correctly? Probably your device is 100% fine.

So it is basically 'your system' (pc hardware, usb driver, operating system, etc) which etc.

I'd try a different system (different pc, os, and cable, maybe at neighbor's or friend's place), just to make sure that "you can get everything to work as it's supposed to work". Then return to your own system and ask yourself why things work well at your friend's system and not so well on your system.

Also try the upcoming version of dex.

 

[Albert_]

That pretty much settles an issue if a bug can't be reproduced. It would be impossible for a software developer to fix something that doesn't exist and can't be reproduced under various conditions on his pc.

On my pc it's two seconds off, tested several times using two different brand new USB cables. Apparently there's a usb bottleneck or overhead somewhere. Possibly there's a problem with my Vista USB universal drivers that are designed for many different types of usb devices. Could be something with the USB Hub controller and drivers. PC Bios, etc.

I would need to change or modify something and the PCLS record the cut-off time to figure out why it's two seconds off.

3 min test

0:02:52 4.108 0.5 23 77.3
0:02:53 4.108 0.5 23 77.3
0:02:54 4.108 0.5 23 77.3
0:02:55 4.107 0.5 23 77.3
0:02:56 4.107 0.5 24 77.3
0:02:57 4.107 0.5 24 77.3
0:02:58 4.107 0.5 24 77.3 (last long entry)

------------------------------

When troubleshooting application related issues you want to first rule out any hardware, driver, connection, cable and OS related problems. This can become time consuming perhaps requiring alot of trial and error. Require various utilities, going into areas of the OS and PC hard-wares not familiar, require some patient and learning. If not willing to learn more about certain issues, as each may be new and different, will make diagnosing problems very difficult if not impossible.

As kreisl stated, another method to troubleshoot is to try running the software application on other PCs or perhaps a PC with a freshly installed OS.

 

[mrQQ]

So it's related to the issue I've mentioned of jumping a second. Here's how it goes:

Tick# - PCLS TimeDisplayed - GraphTime Logged
1 - 00:00:00 - 00:00:00
2 - 00:00:01 - 00:00:01
3 - 00:00:02 - 00:00:02
4 - 00:00:04 - 00:00:03

As I've mentioned, the time displayed on PCLS JUMPS a second randomly, twice a minute, eg from 00:00:02 to 00:00:04, while the graph/csv still counts each second. So that is where the difference comes from.

I've tried all the USB ports and it is the same. In fact, I have no idea how could this be related to USB, if it's clearly an issue with the software. Unless it's pulling the time it displays from the device, but that can only be confirmed by the developers.

 

[chemEJoe]

Noisy fan?

when it's a few hours it did not work, it made that noise at startup for about 20-25 seconds. First batch Stock charger without mod.

I think I'm better to order a new fan before playing with it.

https://www.dropbox.com/s/ka3l6c3agttxucf/Vidéo 16-02-27 20 48 16.mov?dl=0

Hi gyzmo - I am now having grinding fan noises. It was continuous and I had to abort a run. Then I set the fan to ON, and it just grinds away. I tapped it and it seems to be quieter.

Did yours continue? I have not modded anything, nor have I opened the unit. Just seeing where this thread went, and what the conclusion was before I contact GB. I have the fan blowing downward to the tabletop.

 

[chemEJoe]

Inbound QC of 12 NCR18650B Protected from GB, reviewed here by HKJ:

http://lygte-info.dk/review/batteries2012/Panasonic NCR18650B Protected 3400mAh (Green) UK.html

1 was a dud with corroded ends, should have been caught by outbound QC. No voltage, so MC3000 gets no credit Probably the protection circuit is bad, and the cell is ok. I may take it apart to play with it to end up with an unprotected cell if the innards are good.

11 were good, with 5 measured D capacities at:

Disch Capacity Avg: 3286mAh, 22.4 Std Dev
(I used 0.9A to discharge, 0.06A termination).

IR Avg: 102, Std Dev 11.7 (with heavy adjusting of battery seating)

Note IR was all over the map depending on grease on rails and setup (had to move batteries around, start, see of it was reasonable as some were >150mOhm) So my thinking is that IR is probably not going to be too accurate +/- 50mohms as there is contact resistance from different setups. Tell me if I'm missing something.

The rest I could not catch the D cycle screen on time, but the Charge Capacities were close so all 11 good cells are ready to go.

Please let me know what other things I could have learned, thanks in advance.

 

[gyzmo2002]

The rest I could not catch the D cycle screen on time, but the Charge Capacities were close so all 11 good cells are ready to go.

Check post 1765.

At the end of the second charge on a refresh C>D>C you have the discharge capacity on the lcd with fw:1.02.

If you have iOS application, you have the charge capacity at the end.

https://www.candlepowerforums.com/threads/389732

Post 273

 

[gyzmo2002]

Hi gyzmo - I am now having grinding fan noises. It was continuous and I had to abort a run. Then I set the fan to ON, and it just grinds away. I tapped it and it seems to be quieter.

Did yours continue? I have not modded anything, nor have I opened the unit. Just seeing where this thread went, and what the conclusion was before I contact GB. I have the fan blowing downward to the tabletop.

Yes the noise is still there. I did not open the charger to check it. I am waiting this set of 5 fans before open it in case of the original need to be changed.

http://s.aliexpress.com/2mM3yMFf
5pcs lot GDT 12V 2Pin 3007 30*30*7mm 30mm DC Cooler Motor Cooling Micro Processor Fan

 

[chemEJoe]

Check post 1765.

At the end of the second charge on a refresh C>D>C you have the discharge capacity on the lcd with fw:1.02.

If you have iOS application, you have the charge capacity at the end.

https://www.candlepowerforums.com/threads/389732

Post 273

Thanks, gyz ... I thought I read something like that. I'm using 2.8 Android and it matches the 1.02 screen (for Cycle, and I haven't used Refresh). I should have used Refresh - I was using cycle, so that I can just top up the charge using a Charge program, then go straight to D>C to save that first rest time (because I was also playing with different settings and restarting the cycles.) Now I know thank you. Guess it's all going to change when it's time to upgrade to 1.04.

 

[chemEJoe]

Yes the noise is still there. I did not open the charger to check it. I am waiting this set of 5 fans before open it in case of the original need to be changed.

Well, gyz, I was forced to open up the charger today because of that fan grinding noise that just won't go away. I thought well, I can ticket/talk with GB for a week and negotiate a new unit, but it's really something misaligned somewhere, so I took it apart.

It looks promising to fix, as the downward blowing fan should have had both clips engaged. On mine, it's crooked and only one side is engaged. So, the angle might be causing the fan blade to hit, so I think Kreisl's fan mod will fix all this. You may have the same manuf mistake, as they do use a lot of glue.

Since I have it apart, I will do the fan mod according to what Kreisl had posted earlier:

1. clean grease to stabilize the IR readings

2. will smooth out the edges of the Sliders just so it feels better to the touch (Kreisl, I will use my Dremel because the metal looks pretty hard, and I want to do the fan bracket chamfer with it too)

3. will look around for some thermal grease to add to the Slot thermocouples

4. reverse the fan of course, with the spacer as Kreisl advised to avoid fan rubbing

Anything else people remember needs doing?

===

So, does anyone know what the difference with PCB 1.4 vs 1.5 is? I have 1.4 - I ordered my MC just the same week of the switchover oh well. My version does not have a squirrel ... it's some kind of other character. I'd send a pic, but I'm not signed up to any vid or picture sharing sites.

 

[kreisl]

(Kreisl, I will use my Dremel because the metal looks pretty hard, and I want to do the fan bracket chamfer with it too)

Hi, using a file or a dremel for the chamfering is not a good idea. It is a cleaner job with a utility knife.
Difference between 1.4 and 1.5 is the addition of coils near the USB/Bluetooth output to pass FCC/CE certification. It's the only difference. These additional electronic parts may (or may not) even have adverse effects on the accuracy performance of the charger, all original specs were valid for the non-FCC/CE compliant production units like mine and yours, so don't wollie.
I could get my hands dirty on a complimentary 1.5 i guess but oh please :tired: i don't care about the added certification.

While my PCB looked 99% clean, i had used a Q-tip with a drop of solvent/alcohol to make it 100% clean. Nothing else i remember doing. The most important part for us was the fan reversal. I edited my fan reversal guide a day ago, so have a re-check

 

[gyzmo2002]

I have just disassembled the charger. Can I fix this with a little screw? Or I am better to try epoxy glue?


hebergeur dimage

 

[millguy]

I have just disassembled the charger. Can I fix this with a little screw? Or I am better to try epoxy glue?


hebergeur dimage

That happened on one of mine too. In a hurry to get it back together I hooked the end of the spring in the hole and its been fine for over a month.

 

[gyzmo2002]

I think I will try it. If I glue the plastic pole, I will have to disassemble it again. I don't want to play with this mess again.


hebergeur dimage

 

[gyzmo2002]

No more noisy fan. I cut the grill and reversed it. The max temp is 60 celsius 4x1A discharge in a 21celsius room. A good improvement. For the broken pole, I did what millguy told but I put the plastic pole in the hole to prevent it from moving.







No squirrel on the board. V1.4. 2015-07-07. First batch charger.

[FONT=Verdana, Arial, Tahoma, Calibri, Geneva, sans-serif][/FONT]
[FONT=Verdana, Arial, Tahoma, Calibri, Geneva, sans-serif]hebergement image[/FONT]

Télécharger des photos

 

[billcushman]

No more noisy fan. I cut the grill and reversed it. The max temp is 60 celsius 4x1A discharge in a 21celsius room. A good improvement. For the broken pole, I did what millguy told but I put the plastic pole in the hole to prevent it from moving.

Glad you got everything working again. You certainly didn't deserve all the problems you experienced. Hopefully your experience with the MC3000 will be pleasant from now on.

 

[kreisl]

i had mentioned it before but a photo helps:

during reassembly one needs to make sure that the tip of the spring lies in the slot of the slider. this seems obvious but in practice it's not, i know it, and the photo also proves it. it is THE ONLY reason why some slot produces a (faint) screechy sound. the spring itself cannot produce any sound and it also doesn't scratch against any other part. the only scratching part IS the tip of the spring, if it rides on the rail instead of in the slit.

yes, reassembly of the device is straight-forward but challenging to put into practice.