RuiDeng Riden RD6006 DC power supply

[Maxie]

I haven't experienced such thing yet.
However, the longest period I've left it on was like 30 hours on low power settings.
I'm using Unisoft FW.
I wonder if it has anything to do with heat. Have you used high power settings? Which function?
Can you replicate the issue reliably and tell us how to try? Just to verify if its a FW, HW bug or a defect in your unit only.

[ShayBox]

I've noticed it running a 12-15v ~3a load for the past month, it's not consistent, unsure if it happened with stock, honestly doesn't happen enough, it continues to output anyway so I just notice and restart it

[xury]

Hello everybody. This is my fist post on this board.
First of all, I would like to thank Unisoft for the awesome modified firmware.
There are probably many like me who do not need an android application for RD60xx device.
So I decided to completely abandon the mobile application. For me this app is completely useless.
So I used esp-link serial bridge fw and uploaded to wifi dongle.
Now I'm using the RD6018 with  based on msillano node-red flow  and also with the Riden Power Supply application on my PC using virtual com port from https://www.hw-group.com/software/hw-vsp3-virtual-serial-port#Settings.
So now I can manage it from Domoticz and HomeAssistant and also getting whole data from RD6018 to Influxdb and show it using Grafana.
If there is more interest in this solution, when I have some free time I will try to write a complete tutorial about it.

[ShayBox]

So I used esp-link serial bridge fw and uploaded to wifi dongle.
So now I can manage it from Domoticz and HomeAssistant and also getting whole data from RD6018 to Influxdb and show it using Grafana.

In-case you didn't know, and may be interested, you can also use esp-home with the wifi module!
This is my config

[Pukker]

In-case you didn't know, and may be interested, you can also use esp-home with the wifi module!

Yes, sounds interesting, because Ruiden software is useless.
When you can post some information, welcome.

[ShayBox]

Yes, sounds interesting, because Ruiden software is useless.
When you can post some information, welcome.

It's the same as esp-link flashing here
You just flash it with esp-home instead, or something else

[AnotherAlex]

I recently purchased a Riden RD6018 power supply.  The unit came assembled.  The front panel is loose in the housing.  I can gently pull on the encoder knob or any of the posts and the whole panel moves a few milimeters.  I know it uses a snap fit attachment method so not sure what I can do to tighten it up. 

I am curious if they are all this way or if I got one that just doesn't fit tightly.  Are other units the same way with a loose fitting front panel?

Thanks.

--

Edit:  Looks like I found the issue.  I disassembled the unit to put the battery in and noticed that two of the eight screws for the cover are slightly longer than the others.  I moved these screws to the front top and they are long enough to make contact with the plastic of the front panel.  This has tightened up the front panel and it no longer moves. 

[ygi]

Oh so that's how it should be. All the cover screws I got in the kit are identical though. I used some electrical tape over the metal edge to give it the extra tightness so the panel won't rattle anymore. Meh, works well enough.

[squirrel_41]

Hi can I get a copy of the wifi binery thanks.

[UniSoft]

I've noticed from using my 6018 24/7 that sometimes it crashes, the UI freezes and no buttons work, not even power, it'll just get stuck and need the rear power switch turned off. 

Yes, I can confirm.
The problem is localized (at least I hope).
Random freezing due to error in the Modbus driver, so it happens only during communication. 
Fixed, testing now... My RD6006P already working more than 40 hours, connected to PC.
As well, improved the response time (PC software some times loses connection).

While I was looking for the cause of freezing.
I had to implement something like WatchDog timer and BSOD.
If for 20 seconds the WatchDog timer will not be reset, it will turn off the output and goes to BSOD.
If the "AutoRstOnCrash" option is enabled in the menu,
then after another 5 seconds there will be an automatic restart (pressing any button stops the count).
But I'm not sure if this option is needed at all... leave or delete?

Implemented new command for reading dump of the screen,
which uses some kind of packing data.
Now reading the screen dump takes about 2 seconds (at a speed of 115200).

Maybe someone have any other ideas?
Let me know before the release of the new version.

[ShayBox]

Wow that's over and above, I didn't even think about modbus, but yeah my wifi module polls all the time for home assistant, so that makes sense. Thank you for making this amazing firmware with all these features.

[ranseyer]

Hi,
is support for the 6024 expected in the UniSoft Firmware ?

(I will receive the device in ~4 Weeks an i can try to extract infos out of the device if needed...)

[UniSoft]

is support for the 6024 expected in the UniSoft Firmware ?

PS: No custom firmware is planned for RD6024, RD6012P or any further models.
I do not have it.

[GeminiServer]

is support for the 6024 expected in the UniSoft Firmware ?

PS: No custom firmware is planned for RD6024, RD6012P or any further models.
I do not have it.

In General, i love your work and the time you spend into developing such great features for the device, without getting any fee for it.

--> "I do not have it."
Does this mean, when we provide the models to you, you would do the custom firmware for them. Especially the RD6024?
Or collect the amount form the community for you so that you can buy it for yourself? I can also try to contact Ruideng Technology, may they will just send you some examples. 

[xury]

If the "AutoStonCrash" option is enabled in the menu,
then after another 5 seconds there will be an automatic restart (pressing any button stops the count).
But I'm not sure if this option is needed at all... leave or delete?

Hi UniSoft in my point of view is a very important option.
It will by possible to get info about watchdog trigger after restart using modbus?
For example it will by nice to send a message to smartphone about exception.

[Maxie]

BSOD - Leave it in there, disabled as default. I won't be using it, those numbers mean nothing to me.

Watchdog - YES, enabled by default. With an option to switch OFF output, or even reboot the device. There should be some recognizable beep tone as well, like S O S or something.

Interesting to decide when rebooting if the PS should resume operational state as before,
even when my choice is stay OFF when Primary PS powered on, and Output OFF after power on.

I use my PS to charge batteries very often, all kind of chemistries. (This is why I asked for dT/dt)
I like to be on the safe side, and use all available protections combined, Temp, mAh, time, -dV/dt, even min mA.
I don't leave my home during those hours, but I leave the rig unattended.
I enclose batteries in glass and metal containers with ventillation against fire or explosion.
Since all those protections above are implemented in SW, it is an important feature to have a watchdog.
I would use the loud tone to turn my attention to the charging station.

About the non-supported models. I would chip in a couple of bucks. Ppl with new models could chip in more.
If everyone would do it, it would cover procurement cost and have a little bit left over.
If the funds collected were unsufficient, I still wouldn't mind supporting, this FW gives me much joy over the factory one.

[UniSoft]

--> "I do not have it."
Does this mean, when we provide the models to you, you would do the custom firmware for them. Especially the RD6024?

No, it just means that I do not have it.
I just do not want to spend time.

It will by possible to get info about watchdog trigger after restart using modbus?
For example it will by nice to send a message to smartphone about exception.

I didn't implement it... smartphone app doesn't know about custom firmware...

[xury]

I meant reading the info about watchdog trigger event  via modbus and sending this information through other ways. For example, via telegram API.
It seems there are some vacant addresses that could be used for additional functions for eq 0x81 to 0x99.

[Niwivan]

I've noticed from using my 6018 24/7 that sometimes it crashes, the UI freezes and no buttons work, not even power, it'll just get stuck and need the rear power switch turned off. 

Maybe someone have any other ideas?
Let me know before the release of the new version.

Some updates...
https://mega.nz/file/9lMVha7A#KJNN_WOBqJyCZf8KIIIZF8nv3Ys86f4i8T4Eu240hYE

RD60062_V1.36.1e.bin (RD6006)
RD60065_V1.40.1e.bin (RD6006P)
RD60121_V1.34.1e.bin (RD6012)
RD60181_V1.36.1e.bin (RD6018)

Hi,
I'd like to suggest a firmware upgrade. You should be given the option to choose the background color while setting the values. Purple is ugly. See attached image. Thanks Niwivan

Hi i suggest you this color implementation !! Thankss

[UniSoft]

I'd like to suggest a firmware upgrade. You should be given the option to choose the background color while setting the values. Purple is ugly. See attached image. Thanks Niwivan

That background color depends from selected foreground color...
So it is not the one single option, so it is useless to make it changeable.

Code: [Select]

const uint16_t g_colors[16] = {
  0,            // 0 - Default
  C_RED,        // 1  = 0xF800
  C_GREEN,      // 2  = 0x07E0
  C_BLUE,       // 3  = 0x001F
  C_WHITE,      // 4  = 0xFFFF
  C_YELLOW,     // 5  = 0xFFE0
  C_MAGENTA,    // 6  = 0xF81F
  C_CYAN,       // 7  = 0x07FF
  //
  C_LIGHT_BLUE, // 8  = 0x01CF
  C_GRAY,       // 9  = 0x8410
  C_BROWN,      // 10 = 0xA145
  C_ORANGE,     // 11 = 0xFD20
  C_OLIVE,      // 12 = 0x8400
  C_TEAL,       // 13 = 0x0410
  C_PINK,       // 14 = 0xFDF9
  C_MAROON      // 15 = 0x8000
};

const uint16_t g_selcolors[16] = {
  C_MAGENTA,    // 0
  C_AQUA,       // 1
  C_MAGENTA,    // 2
  C_MAGENTA,    // 3
  C_MAGENTA,    // 4
  C_MAGENTA,    // 5
  C_AQUA,       // 6
  C_MAGENTA,    // 7
  C_WHITE,      // 8
  C_BROWN,      // 9
  C_AQUA,       // 10
  C_BLUE,       // 11
  C_BLUE,       // 12
  C_ORANGE,     // 13
  C_MAGENTA,    // 14
  C_AQUA        // 15
};

[tomvid]

Hello UniSoft, thank you so much for your great job for our Ridens!

Working with 1.37.1f on my RD6006 I have found one bug regarding the "SaveGraphSett" function: it works for the setting of voltage, current, temperature, but NOT for the wattage, as you can see below in the clip I just recorded:



As you can see, my graph setting for the Power curve was set to "OFF" (time 0:11). However, after rebooting, the setting goes itself to "0.01W" (time 0:20), so I need to set it again manually (time 0:25). It is very easy to reproduce this bug, since it occurs always for me (time 0:31 and later 1:02).
Would it be possible to correct this bug? I use pretty often the charging function, so it would be great, to have a patch here.




And the 2nd topic: I'm very happy with the "CutOff -ΔV/Δt" since even with the smallest setting 10 mV it works really reliable on my system when charging a single AA cell. From the manufacturer of the cells (Panasonic) I know, the optimal threshold is 7,5 mV per single cell, but of course I know the Riden's hardware does not have that good resolution, so we need to take 10 mV (setting 0.01 V) which is still pretty good, as I tested it on more than 20 cells.

UniSoft, would it be possible to add an adjustable time delay before arming the "CutOff -ΔV/Δt"? I know in the current firmware this value is hard-set to 5 minutes, so the the cutoff is "armed" after the first 5 minutes. My problem is, some of my cells need up to 40 minutes for the voltage rise. This means, the voltage goes down for the first 40 minutes since cells are getting warmer. This triggers the CutOff far to early for me. My solution is to deactivate "CutOff -ΔV/Δt" for the first 20-40 minutes and after I see on the graph, the voltage rises, I activate CutOff manually. Finally the voltage drops again and "CutOff -ΔV/Δt" terminates the charging. But, as I said, this is a manual process. If I only could postpone the arming of "CutOff -ΔV/Δt" by 30 minutes, this could work really autonomously. Since the time delay may be cell-specific I think the adjustable time delay would be the best for all of us. With the default value of 5 minutes it would be exactly the same as it is now. But having it adjustable would help for some older cells that some users have.

[UniSoft]

I have found one bug regarding the "SaveGraphSett" function: it works for the setting of voltage, current, temperature, but NOT for the wattage...

Yes confirm.
It is not working for all options, if you select OFF while Y offset is 0.
I will fix it.

"CutOff -ΔV/Δt" since even with the smallest setting 10 mV...

Actually we can try to make it less... for example starting from 0.005V... with step 0.001V

would it be possible to add an adjustable time delay before arming the "CutOff -ΔV/Δt"?

Possible, why not?
What name for option (max 14 characters)? what range, step?

Here is how it is checking
 

Code: [Select]

          if (g_dVdTCur)
          {
            // Check for -dv/dt condition
            // Minimum time before termination in DVDT mode is 5 minutes
            if ((tickCount - g_StartTimeStamp) > (5 * 60 * 1000) && (g_outVoltageFiltered + g_dVdTCur) < g_outVoltageFilteredPeak)
            {
              chgstate = CHARGE_OFF_DVDT;
            }
            // Check for 0dv/dt condition (Default is 15 minutes)
            else if ((tickCount - g_PeekVoltageTimeStamp) > (15 * 60 * 1000))
            {
              chgstate = CHARGE_OFF_0DVDT;
            }
          }

[tomvid]

Hello UniSoft, I really do appreciate your openness!

1. Step value for the voltage "CutOff -ΔV/Δt":
I like your idea with starting with 5 mV and the step of 1 mV. Maybe it is even more convienient to use [mV] as the main unit for this setting instead of [V].



2. Delay for CutOff -ΔV/Δt here are my proposals:

• name: "-ΔV/Δt Lockout"   OR    "CO-ΔV/Δt Delay"     OR    "-ΔV/Δt Delay"
• step: 1 minute
• range: 1 - 60 minutes

However, correct me if I'am wrong, I think we need also to disable the 0dVdt termination for the lockout time defined for dVdT, since the 0dVdT will trigger after 15 minutes first, because no new PeekVoltageTimeStamp will be found.

Example:
if the new parameter "-ΔV/Δt Lockout" is set to 30 minutes, then the 0dVdT detection will terminate the charging after 15 minutes with 0dVdT, because PeekVoltageTimeStamp will not change within 15 minutes.

So I think, this extra condition for the 0dVdT is needed:

Code: [Select]

(tickCount - g_StartTimeStamp) > (lockout_time * 60 * 1000)


[arekm]

Did anyone have a problem where lcd behaved normally in past usages and now it is white only (backlight is on, nothing else is displayed) ? Is it LCD fault or something else?

Reset (0 + power on) doesn't help. Firmware upgrade also doesn't help.

(note module itself mostly works, just not screen; tested via usb & pc software)

[ShayBox]

Sounds like using the wrong file for your device, there's two versions of some of the files because Riden changed the screen mid production