Author Topic: EEZ Bench Box 3 - Sequel to EEZ H24005  (Read 151327 times)

0 Members and 1 Guest are viewing this topic.

Offline Ranayna

  • Frequent Contributor
  • **
  • Posts: 856
  • Country: de
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #450 on: October 21, 2020, 02:18:15 pm »
Ok, if this is a likely fix, i will throw in an additional (or two) LT3763 when i finally manage to order the parts for my bare boards kit. :)
Thanks for the quick reply.
 

Offline fuzzoli

  • Regular Contributor
  • *
  • Posts: 70
  • Country: us
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #451 on: October 21, 2020, 02:19:59 pm »
...any thoughts on to what might be wrong?
All is good -- apparently it was a PEBCAT issue (Problem Existed Between Chair and Touchscreen).  |O  :palm:

The system was powering up with the current limits set to 0.  Once setting a non-zero current limit, everything (including the coupling) is working great!

Thanks for all the suggestions.

 
The following users thanked this post: Kean, prasimix, Andrew McNamara

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #452 on: October 21, 2020, 02:21:44 pm »
Ok, if this is a likely fix, i will throw in an additional (or two) LT3763 when i finally manage to order the parts for my bare boards kit. :)
Thanks for the quick reply.

Many thanks!

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #453 on: October 21, 2020, 02:23:10 pm »
...any thoughts on to what might be wrong?
All is good -- apparently it was a PEBCAT issue (Problem Existed Between Chair and Touchscreen).  |O  :palm:

The system was powering up with the current limits set to 0.  Once setting a non-zero current limit, everything (including the coupling) is working great!

Thanks for all the suggestions.

 :-+  :-/O

Offline danielbriggs

  • Regular Contributor
  • *
  • Posts: 127
  • Country: gb
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #454 on: October 21, 2020, 02:44:03 pm »
Just another small update:

- Large plain washers added to screen studs. Now the one that was off-centre is perfect + solid. Would reccomend adding washers as give much more adjustment play + stable mounting of the display.

- Screws swapped out for black stainless ones; should make changing modules in the future better  :-+

- Earth'd the two extra pannels as per one members good suggestion. I used the rear PSU mounting post + removed the paint carefully on the corresponding screw hole + hole by the IEC connector inlet. Crimped some earth wires to each and all pannels tested well.

I've done all of the above to both units, so they're now absolutely perfect. Really happy with these overall. :-DMM

All the best,
Dan
 
The following users thanked this post: Kean, prasimix, mcdanlj

Offline jhenderson0107

  • Regular Contributor
  • *
  • Posts: 133
  • Country: us
    • Elk Engineering
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #455 on: October 21, 2020, 03:49:40 pm »
prasimix -

I'd like to return them for repair.  I don't have the time to diagnose and rework them myself.  Please PM shipping instructions. 
 

Offline Andrew McNamara

  • Regular Contributor
  • *
  • Posts: 76
  • Country: au
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #456 on: October 21, 2020, 11:45:55 pm »
Installing either module individually or both simultaneously, module self-test will fail frequently (~50%).  Both modules fail sporadically and with identical symptoms.  Typically, only a single module fails during a boot session.  Rarely, both modules will pass self-test.  If a module passes self-test, it can successfully drive 1A @ 20V into a Maynuo 9811 active load.  If a module fails self-test, it cannot be used unless the system is power-cycled and the module happens to self-test successfully. 

It does seem odd that both modules exhibit exactly the same fault... it makes me wonder about common elements... maybe a comms issue on the backplane or between the backplane and MCU board, or maybe there is an issue with the Aux Power supply board - either intermittent comms with the board, or a fault in the soft start/standby circuit? You might want to check the cable that runs from the MCU to the Aux power supply board - a number of people have reported that it has been crimped at an angle and has shorts (but this could also exhibit as intermittent comms). If the PWR_DIRECT signal wasn't making it to the Aux Power board, I could imagine might manifesting this way (the power supplies would come up too slowly). If poking around this board, just remember that much of it has exposed mains voltages.

Maybe the firmware should have an extended diagnostic test option where verifies comms is stable?
« Last Edit: October 22, 2020, 12:16:40 am by Andrew McNamara »
 
The following users thanked this post: Kean, jhenderson0107, AlanS

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #457 on: October 22, 2020, 07:30:30 am »
The error messages in the log file show quite some intermittent problems with the HW connected via SPI. A possible issue could be just the speed of the isolators. Unless absolutely needed the FW could try to reduce the SPI clock a little.  At least slow isolators reduce the chance of errors due to clock ringing. It could still be some borderline signal quality or timing.
The different reaction of the FW versions may indicate a timing problem.
So it is not clear if the problem is with the CPU board, the back-plane (less likely) or the modules.

The log file is a really nice feature for debugging  :-+
 
The following users thanked this post: prasimix

Offline Ranayna

  • Frequent Contributor
  • **
  • Posts: 856
  • Country: de
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #458 on: October 22, 2020, 08:56:52 am »
To debug a module, is it possible to extend the connection to the BPC with a ribbon cable? Or would i already run into signal integrity issues?
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #459 on: October 22, 2020, 10:21:00 am »
A cable extension should ideally work, if not excessively long. If not it would be a sign of a relatively sensitive design.  Ideally I2C chips should ignore fast ringing as they don't have to be that fast. However it depends on the chips.
The isolator chips used (SI8641) seems to be quite fast, but it should still suppress really fast ringing. Slower isolators may be an option (if available) to get less critical. I don't think many modules would need full speed and the SPI bus should be OK with a destination specific speed. A lower speed mode may also help the relay cards to get away with a lower clock µC.
 

Offline jhenderson0107

  • Regular Contributor
  • *
  • Posts: 133
  • Country: us
    • Elk Engineering
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #460 on: October 22, 2020, 01:15:33 pm »
All - Thanks for responding. 

@Andrew McNamara: I agree that the common failure mode suggests a problem upstream of the modules.  I will crimp another cable assembly together and give it a try. 

@Kleinstein: I have not yet tried the 3rd module site - I'll do that.  But swapping modules in each of the first two slots does not change the failure mode.  So perhaps the root cause is upstream at the CPU or some key cable. 
 

Offline danielbriggs

  • Regular Contributor
  • *
  • Posts: 127
  • Country: gb
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #461 on: October 22, 2020, 01:36:03 pm »
Today I've used both BB3's on actual work powering some LED's and eval boards. Works a treat!  :-+

However: in my playing I noticed if you set the OCP to values lower than 20ms and enabled, strange things happen. i.e. the whole chassis reboots? / enters standby when trying to enable the module.

For DCP405's... setting the value to less than 3ms = causes the stanby bug on output enable.
For DCM220's... setting to less than 10ms = causes the standby bug on output enable.

If it is out of spec to have the OC protection work <20ms on modules (the default). This is ok, but I think the user should be prevented from trying to enter "invalid" values. If the units are designed to work down to "0ms", then any ideas what is causing the sudden stanby panics?


I made two quick vids showing:

First vid = Ch1 @ 10ms, Ch2 @ 1ms (both DCP405). http://danielbriggs.co.uk/images/BB3_OCP_Issue1.mp4
Second vid = Ch2 @ 10ms, Ch2 @ 5ms (both DCM220). http://danielbriggs.co.uk/images/BB3_OCP_Issue2.mp4
 

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #462 on: October 22, 2020, 01:40:26 pm »
Today I've used both BB3's on actual work powering some LED's and eval boards. Works a treat!  :-+

However: in my playing I noticed if you set the OCP to values lower than 20ms and enabled, strange things happen. i.e. the whole chassis reboots? / enters standby when trying to enable the module.

For DCP405's... setting the value to less than 3ms = causes the stanby bug on output enable.
For DCM220's... setting to less than 10ms = causes the standby bug on output enable.

If it is out of spec to have the OC protection work <20ms on modules (the default). This is ok, but I think the user should be prevented from trying to enter "invalid" values. If the units are designed to work down to "0ms", then any ideas what is causing the sudden stanby panics?


I made two quick vids showing:

First vid = Ch1 @ 10ms, Ch2 @ 1ms (both DCP405). http://danielbriggs.co.uk/images/BB3_OCP_Issue1.mp4
Second vid = Ch2 @ 10ms, Ch2 @ 5ms (both DCM220). http://danielbriggs.co.uk/images/BB3_OCP_Issue2.mp4

Ok, please let me know what you have defined here under "Shutdown when protection tripped":


Offline danielbriggs

  • Regular Contributor
  • *
  • Posts: 127
  • Country: gb
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #463 on: October 22, 2020, 01:45:59 pm »
Perfect! That was it. Thanks for the instant reply.
I didn't know I had that enabled / or it was even a feature. Switched "Shutdown when tripped protection" to "NO" and working well now.

Sorry - user error on my part!  :palm:


Ok, please let me know what you have defined here under "Shutdown when protection tripped":


 
The following users thanked this post: AlanS

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #464 on: October 22, 2020, 01:48:50 pm »
You and others will have to start living with the fact that this device has so many options :)
 
The following users thanked this post: danielbriggs, Kean, gnavigator1007, AlanS

Offline AlanS

  • Supporter
  • ****
  • Posts: 103
  • Country: au
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #465 on: October 23, 2020, 12:15:34 am »
I can just imagine Denis' face after all the travails. Something went VERY right. Good job Daniel for reporting the feature - it helps.
 

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #466 on: October 26, 2020, 11:40:54 am »
New firmware v1.4 is now available for download. It mainly addresses multiple issues reported by the BB3 backers.

    #33 The detected module fault is pushed to standby
    #60 Replace SYSTem:CHANnel:SNO? with SYSTem:SLOT:SNO?
    #61 Record in event log MP script execution start and end
    #67 NTP server connection frequency
    #68 Decrease resolution of manual fan speed control
    #69 "Encoder & User sw" menu improvement
    #70 Inconsistencies in the number of decimal places displayed
    #72 Further encoder functionality improvements
    #73 Numeric value rounding issue
    #74 Disable Down-programmer in low current range
    #76 Lightness issue in Dark mode
    #77 Retain encoder step (via User SW) setting on power up
    #80 DLOG current background highlighting in 2-ch mode
    #81 Encoder step size changing improvement
    #84 Additional check for parallel coupling
    #86 List preview issue
    #87 Encoder step values for voltage and current for DCP405
    #89 Wrong value in low current calibration graph
    #93 Disable User sw menu when Power/Reset control menu is active
    #95 USB is disabled after firmware upgrade from v1.0
    #96 Garbage in event log on power up without SD card
    #97 Current readout in YT-view and dark theme could be more readable

Download procedure is described in Chapter 13.
 
The following users thanked this post: danielbriggs, exe, gnavigator1007, AlanS, Andrew McNamara

Offline danielbriggs

  • Regular Contributor
  • *
  • Posts: 127
  • Country: gb
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #467 on: October 26, 2020, 12:24:09 pm »
Hi Denis,
Thanks for an awesome update. Installed onto both units sucessfully just now.
(Appriciate the new OS X upgrade instructions; works really well).  :-+

Very very very minor requests:

- Is it possible to get the built in BB3 file manager to hide hidden folders?
When you plug one of the SD cards into a Windows PC you get "System Volume Information" folder created. And plugged into OS X you get lots of crap: ".Trashes", ".fseventsd", ".Spotlight...." etc.
It would just offer less clutter to the nice file manager view you have made.

- I'm sure it's a rounding fragment, or floating point issue, but often when the display is reading "0V" or "0A" it's displaying as "-0.0V or "-0.0A" this seems a bit illogical, so any way to clamp it to remove the minus when it's at zero (or as close as possible). Or am I being too picky  :scared: (pls tell me to shut up if so!)

Keep up the good work.
All the best,
Dan
 
The following users thanked this post: AlanS

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #468 on: October 26, 2020, 01:01:03 pm »
Very very very minor requests:

- Is it possible to get the built in BB3 file manager to hide hidden folders?
When you plug one of the SD cards into a Windows PC you get "System Volume Information" folder created. And plugged into OS X you get lots of crap: ".Trashes", ".fseventsd", ".Spotlight...." etc.
It would just offer less clutter to the nice file manager view you have made.

Hi Dan, feel free to open a new issue with for this on Github.

- I'm sure it's a rounding fragment, or floating point issue, but often when the display is reading "0V" or "0A" it's displaying as "-0.0V or "-0.0A" this seems a bit illogical, so any way to clamp it to remove the minus when it's at zero (or as close as possible). Or am I being too picky  :scared: (pls tell me to shut up if so!)

Yes it is a rounding issue. Try to open issue for this, and we'll see what Martin will tell us :)

Offline knotlogic

  • Regular Contributor
  • *
  • Posts: 189
  • Country: sg
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #469 on: October 26, 2020, 02:02:24 pm »
I have a question about grounding:  I know we can remove a little paint around the mounting hole on the bottom plate where it meets the grounded  mount point on the AUX-PS module.  But is there any benefit to removing the paint around the mounting holes where the MCU and BP3C modules are mounted?
 

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #470 on: October 26, 2020, 02:49:58 pm »
I have a question about grounding:  I know we can remove a little paint around the mounting hole on the bottom plate where it meets the grounded  mount point on the AUX-PS module.  But is there any benefit to removing the paint around the mounting holes where the MCU and BP3C modules are mounted?

Not at all. Metal stand-off on that two modules goes into isolated holes.

Offline danielbriggs

  • Regular Contributor
  • *
  • Posts: 127
  • Country: gb
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #471 on: October 26, 2020, 03:01:45 pm »
Hi Dan, feel free to open a new issue with for this on Github.

No problem; I've shuffled those two into GitHub.

One other thing I forgot to mention: the encoder changes in V1.4 seem 10x more sensitive than they were in any other version, this is no bad thing, but just a heads up for anyone else noticing the different feel. I had to change the acceleration to Up: 1, Down: 3, to make it "feel" right for me. The default of U:6, D:8 was crazy fast, at least for the things I work with.

All the best,
Dan
 
The following users thanked this post: prasimix, AlanS

Offline jhenderson0107

  • Regular Contributor
  • *
  • Posts: 133
  • Country: us
    • Elk Engineering
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #472 on: October 26, 2020, 03:04:09 pm »
prasimix -

Returning to the issues I've reported:  Is it still your position that I must return both modules for repair?  If so, please PM instructions. 
 

Offline Andrew McNamara

  • Regular Contributor
  • *
  • Posts: 76
  • Country: au
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #473 on: October 27, 2020, 08:07:11 am »
Returning to the issues I've reported:  Is it still your position that I must return both modules for repair?

My concern with sending them back is that the issue may lie elsewhere - it would be frustrating and disappointing to get new modules and find you still have the same problem.

I haven't seen any other people reporting this problem, and it seems unlikely you would have received two modules that tested okay for Denis, but developed identical subtle faults during transit - my money would be on some other common element. Unfortunately, this is one of the risks of buying a kit of modules - the modules have been tested, but not tested together. The errors suggest to me intermittent I2C comms with the DCP405 modules, yet other I2C modules in the system appear okay, eg fan controller, and backplane, so maybe the issue is with the power supplied to the DCP405 modules. Since the DCP405 have independent Meanwell 48V power supplies - maybe there is an issue with their upstream supply from the AUX board (maybe the soft start), or a loose AC connector?

BTW, I'm just another customer - I don't speak for the Envox team.
 
The following users thanked this post: prasimix

Offline prasimix

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: EEZ Bench Box 3 - Sequel to EEZ H24005
« Reply #474 on: October 27, 2020, 08:14:18 am »
Returning to the issues I've reported:  Is it still your position that I must return both modules for repair?

My concern with sending them back is that the issue may lie elsewhere - it would be frustrating and disappointing to get new modules and find you still have the same problem.

I haven't seen any other people reporting this problem, and it seems unlikely you would have received two modules that tested okay for Denis, but developed identical subtle faults during transit - my money would be on some other common element. Unfortunately, this is one of the risks of buying a kit of modules - the modules have been tested, but not tested together. The errors suggest to me intermittent I2C comms with the DCP405 modules, yet other I2C modules in the system appear okay, eg fan controller, and backplane, so maybe the issue is with the power supplied to the DCP405 modules. Since the DCP405 have independent Meanwell 48V power supplies - maybe there is an issue with their upstream supply from the AUX board (maybe the soft start), or a loose AC connector?

BTW, I'm just another customer - I don't speak for the Envox team.

Thanks for you input. What Jim reported is a SPI not an I2C issue. Of course there is still some small chances that something is wrong on the MCU modules. We'll find out that soon.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf