Author Topic: EEZ Bench Box 3 - Sequel to EEZ H24005 (Read 152796 times)

Mike G · « **Reply #625 on:** July 03, 2021, 05:43:43 am »

Ok, more clues

Swapped faulty DCP with good DCP, including swapping Meanwell supply so faulty DCP is using good Meanwell, and fault moved with DCP module.
After more experimentation discovered fault only shows when current limit is set at 51mA or higher. 50mA or lower CC led behaves correctly.
Had a look at circuit diagram to suss out what is going on, all the parts relating to this section are under the heatsink so access very poor for inspection/testing. Not impossible but more thought needed before diving in.
Just to avoid wasting time I thought a recalibration would be a useful thing to check.
Problem... it asks me for a calibration password, I have never set a calibration password.... is there a default one set even though the user manual says not... HELP

If you think the problem with the CC led is indeed around the current sense switching area I am both happy and interested to track it down. I will, however have to make up extension leads to allow working on it outside the case.
It still seems strange this has only shown up after upgrade to v1.7, what do you think about reverting to 1.6.1 or does this not help?
Regards Mike

Mike G · « **Reply #626 on:** July 03, 2021, 05:48:03 am »

Sorry, I had a Muppet moment, read the user manual again and discovered it is the system password that is undefined but the calibration password is...

prasimix · « **Reply #627 on:** July 03, 2021, 06:01:50 am »

You can try that but it cannot be a backplane "issue" since CV and CC are analog indicators, i.e. not under control of MCU so using different SPI bus on different slot will not make any difference.

prasimix · « **Reply #628 on:** July 03, 2021, 06:04:45 am »

The default calibration password is eezbb3 as noted in https://www.envox.eu/eez-bench-box-3/bb3-user-manual/12-power-modules-calibration/

Mike G · « **Reply #629 on:** July 03, 2021, 06:10:34 am »

Yes, I did eventually find that please see my latter post
Anyhow have now recalibrated the dodgy DCP and all seems fine now

Just did two, 50mA and 4.8A, points.
Thank you so much for your time and advice.
Really happy with BB3 and v1.7
Regards Mike

prasimix · « **Reply #630 on:** July 03, 2021, 06:13:18 am »

Ok, glad to hear that you managed to fix that with calibration. it seems that low range (50 mA) required recalibration.

Mike G · « **Reply #631 on:** July 03, 2021, 06:19:46 am »

The fault only showed up above 50mA so I assumed the problem was with the High current calibration.
Anyhow all ok now, seems strange this only showed up after upgrading.
The thing I noticed was that the dodgy DCP had four calibration points when I checked, two of them very low current, less than 50 mA so I erased the two low ones and just recalibrated at 50mA and 4.8A. The DCP that showed no problems only had calibration points at 50mA, 2A and 4.8A so I suspect the problem was caused by having calibration points below 50mA in the high current calibration

prasimix · « **Reply #632 on:** July 03, 2021, 06:23:09 am »

Good, to know. Just be sure that you have at least 4 calibration points in total: two for low range (50 mA) and two for high range (5 A).

Kleinstein · « **Reply #633 on:** July 03, 2021, 08:21:36 am »

The CC led going on could be related to an offset in the set current limit. So if the current limit is set very low, it may actually be set below 0 and than cause the CC mode to engage when off. It should not cause a problem - more like a bit confusing because of the LED, but hardly any change to the output.

Chances are that on turn on of the outout the CC mode will engage anyway for a short time ( ~ ms, so may not be very visible) until the voltage reaches it's set values. I asume the DCP does not copy the R&S supply "bug" in setting high the current limit for a few ms when the voltage is supposed to rise.
The Firmware update may have changes to off part - looks like that in off state the current limit is also set down (not sure if actually needed, maybe in the hope to reduce overshoot in some cases).

prasimix · « **Reply #634 on:** July 03, 2021, 09:50:46 am »

@kleinstein, currently we have the following situation: When power output is enabled, we don't have any overshoot. For "R&S" setup, i.e. Uset=3.3 V and Iset=1 A with 2R load we're entering CC mode like this:

However there is another scenario that Dave didn't tested: a "list programming" (that could be R&S's EasyARB). Here we can overshooting for ~1 ms if Uset is fixed (e.g. 3.3 V) and current is programmed between 20 mA and 1 A when current range is set in BEST:

A note about that is exists in our User manual:

Important: when power module is operating in CC mode, and load is changed dynamically that current vary between low and high current range, switching between ranges will generates voltage overshoots due to finite switching speed. Set current range manually in that case to avoid voltage overshoots.

If we just change current range selection from BEST to HIGH then everything is fine again:

Perhaps we should force current range selection from BEST to HIGH when list programming is selected to avoid this issue.

prasimix · « **Reply #635 on:** July 08, 2021, 04:30:57 pm »

The EEZ BB3 firmware version 1.7.1 has been released. As before, you can find instructions for the firmware upgrade in Chapter 13.

Mike G · « **Reply #636 on:** July 10, 2021, 04:45:06 am »

V1.7.1 working really well.
The function generator is massively improved, now very useable.
True AC output

jklasdf · « **Reply #637 on:** August 30, 2021, 05:00:28 am »

Quote from: prasimix on March 20, 2021, 05:53:13 pm

On Mouser stock is currently Bolek&Lolek production that is shown in Ian's video but also include first DCP405s from new batch made by another PCBA contractor (blue PCB). I've checked and repair all modules sent to Mouser, but some of them still had issues probably due to questionable soldering. New batch is delayed for almost two months due to shortage of components: Mean Well IRM AC/DC converters in first place. DCM module is particularly sensitive since it seems that LT3763 doesn't tolerate questionable thermal pad joint.

Is Mouser's stock of the two channel kits (currently showing 30 in stock) still the old production PCBs with questionable soldering? I've been looking into buying one of these supplies (in the US), and so far Mouser seems like the cheapest option. The main other option seems to be buying from the EEZ web store, which would be newer production, pre-assembled, and with a warranty, but also ends up being more than $120 more expensive shipped to the US (possibly more, haven't looked into fees), at which point it would make sense to just buy the 4 channel version from Mouser (not in stock yet, currently showing 23 on order, which should probably be newer production?). Only other consideration is I do like the thought of supporting EEZ on their web store, but I also doubt they see even a small fraction of that price difference.

prasimix · « **Reply #638 on:** August 30, 2021, 05:41:55 am »

Quote from: jklasdf on August 30, 2021, 05:00:28 am

Quote from: prasimix on March 20, 2021, 05:53:13 pm
On Mouser stock is currently Bolek&Lolek production that is shown in Ian's video but also include first DCP405s from new batch made by another PCBA contractor (blue PCB). I've checked and repair all modules sent to Mouser, but some of them still had issues probably due to questionable soldering. New batch is delayed for almost two months due to shortage of components: Mean Well IRM AC/DC converters in first place. DCM module is particularly sensitive since it seems that LT3763 doesn't tolerate questionable thermal pad joint.

Is Mouser's stock of the two channel kits (currently showing 30 in stock) still the old production PCBs with questionable soldering? I've been looking into buying one of these supplies (in the US), and so far Mouser seems like the cheapest option. The main other option seems to be buying from the EEZ web store, which would be newer production, pre-assembled, and with a warranty, but also ends up being more than $120 more expensive shipped to the US (possibly more, haven't looked into fees), at which point it would make sense to just buy the 4 channel version from Mouser (not in stock yet, currently showing 23 on order, which should probably be newer production?). Only other consideration is I do like the thought of supporting EEZ on their web store, but I also doubt they see even a small fraction of that price difference.

Due to a problem with components shortage we are not able to continue production so it cannot be expected for Mouser to get 4-channel before the second quarter of next year. However, they have enough DCMs in stock (latest version) so it can possibly be combined with a 2-channel kit version. Also, 65 new starter kits (without power modules) were delivered to Mouser long time ago, which are still not visible and I don't know what the idea was with that. Maybe you should contact CS for more information.

jklasdf · « **Reply #639 on:** August 31, 2021, 01:54:28 am »

Sounds frustrating, do you know if Mouser's current stock of the 2-channel version is the older production with possibly bad soldering?

prasimix · « **Reply #640 on:** August 31, 2021, 05:36:18 am »

Quote from: jklasdf on August 31, 2021, 01:54:28 am

Sounds frustrating, do you know if Mouser's current stock of the 2-channel version is the older production with possibly bad soldering?

Highly likely that it is an older production.

prasimix · « **Reply #641 on:** October 04, 2021, 10:24:03 am »

After a lot of complications, the new SMX46, PREL6 and MUX14D modules are now available in our web store.

https://www.envox.eu/product/smx46-module/
https://www.envox.eu/product/prel6-module/
https://www.envox.eu/product/mux14d-module/

prasimix · « **Reply #642 on:** October 19, 2021, 11:55:01 am »

Firmware interim release 1.7.2 is now available for download: https://github.com/eez-open/modular-psu-firmware/releases/tag/1.7.2

prasimix · « **Reply #643 on:** October 21, 2021, 08:12:19 am »

Limited number of MIO168+AFE3 modules are now available in our web store: https://www.envox.eu/product/mio168-afe3/

elekorsi · « **Reply #644 on:** November 08, 2021, 01:52:43 pm »

Unfortunately my BB3 died today

Using it normally in the morning for some quick tests, everything normal. After about an hour or two, i powered it up again and the problem appeared... It powers up, everything seems normal and then after few seconds it puts all channels into error. You can see on the video, exactly how it behaves: https://youtu.be/3RUxj3n_7_g
Visual inspection shows nothing abnormal. I tried with taking out the modules and inserting them one by one in different locations, in case that one of the module would be causing problems. No matter what i do, it is always the same result. What is strange to me, that in those few seconds before the error occurs, modules are fully functional, you can switch them on/off, change voltage, etc. When i switch the channel ON, i can measure the correct voltage on the output, but then it goes to error.

Any idea, what could this be?

It has a MCU board r2B4 with the date of 01/2020
The configuration is 2x DCP405 + 1x DCM224 (ordered and installed this year)

prasimix · « **Reply #645 on:** November 08, 2021, 02:00:00 pm »

Hi, elekorsi, check AC power distribution, it's quite possible that your soft-start control on the AUX-PS failed, and that could be related to issue described in #108 . As a remedy you could try to add bigger RC snubber or replace AUX-PS with the latest revision (both bare PCB and assembled module are available on web store).

prasimix · « **Reply #646 on:** November 08, 2021, 02:01:59 pm »

Forgot to mention, upgrading firmware to the latest revision 1.7.3. is highly recommended (many things are fixed and improved in the meantime).

elekorsi · « **Reply #647 on:** November 08, 2021, 02:30:00 pm »

Hi, yeah, i then remembered that i saw something about the overheating resistor on AUX board and checked that one (R4). I just touched it and it fell OFF, and it sure had some tough life, as it's resistance is around 460ohms instead of 330
As i didn't have a 330ohm SMD resistor in that size, i just made some modern art with the through hole one. Will replace it at next order... And yes, IT WORKS

I will install a RC snubber and will also do the FW update today

prasimix · « **Reply #648 on:** November 08, 2021, 02:32:48 pm »

Some other value should also works e.g. 220 to 360R, but in the long run you need to add bigger RC snubber.

elekorsi · « **Reply #649 on:** November 11, 2021, 12:53:11 pm »

It's me again

I replaced the resistor with proper one, installed a RC snubber and updated the FW.
I noticed a problem on the CH2 today. If i reduce the voltage to quickly it goes into error (3% OV protection). If i go slowly, it is OK. I tried with funtion generator and of course it goes into an error as soon as i start it. Is something wrong with the down programmer?
https://youtu.be/bG2VoespuyM


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: EEZ Bench Box 3 - Sequel to EEZ H24005 (Read 152796 times)

Share me