The disappointing child with Rev 1856 / Rev 1860 brings new issues

[bmrc]

Chapter 1: The disappointing child (1856)

I skipped over the Chinese garbage and spent my hard earned cash on a 'better' charger, or so I thought. After much research the KID seemed to fit my requirements - MPPT, 100v+ PV, 24v battery support, and serial data out. I have a remote off-grid property, where I have a few cameras up to catch nature, and (hopefully never) trespassers/thieves. These cameras run on a 24v, 440ah bank of batteries, 360 watts worth of Solar Panels. I do have slow cellular data so I can remotely connect to a small computer to manage the cameras and view the serial data from the KID.

Installation of the KID was simple and it worked great for a week while I was at the property, cycling through Bulk, Absorb, Float as I would expect. I fumbled around the Interwebs and came up with the wiring I needed to create a serial cable to talk to the KID (Hey, maybe document this in a manual MidNite). Success! Serial data on the PC so I can see the health of the batteries remotely. "This works great!" I thought. I returned to my house/job/reality 6 grueling hours of driving away.

I thought wrong. The KID continued to work for about 5 more weeks, then one day I stopped receiving serial data. I restarted the computer, no change. "Maybe I messed up on the cable or somethings wrong the with the computer (it runs Windows after all), I'll figure it out next time I return". Life went on, and I continued to receive pictures for another few weeks until, nothing. I could no longer remotely connect.

After finding the time to return to my remote property (another 6 grueling hours of driving later) I was greeted with a 24v bank of batteries at 11v, and a KID charger that was FROZEN. No response to any buttons.

<rant>
A CHARGE CONTROLLER THAT DOESNT CHARGE IS F**KING USELESS! Is this a quality, tested product or someones arduino project on instructables?
</rant>


Chapter 2: New Issues (1860)

After a few choice words, I disconnected the KID and hooked up the only small charger I had. I charged the batteries for literally days using the generator and small charger. They did finally come back up in voltage, but I'm sure their lifespan has been impacted - only time will tell. Oh and I burned about 15 gallons of fuel, and drove off any wildlife that don't care for the clickity clackity of a generator (that's all of them).

While waiting for the batteries to charge I fumbled around the Interwebs some more and found KID had a newer firmware available, and whadda you know? In the notes for 1860 it states "Fixes sporadic Kid Freezes and bugs in Aux menu" Since this sounds exactly like the issue I'm having I break out a USB cable and download the new firmware. Once the batteries were charged to at least 24v, I hooked the KID back up and updated the firmware.

Hopefully 1860 fixes the freezing issue. I'ts only been a week since the update, so I can't say for sure yet.

I have already found a new issue with 1860 - The format of the serial data is drastically changed. No documentation, just change sh*t MidNite, that's the way! Not like anyone built custom software to read this data *COUGH COUGH* Again - this feels like an Instructables project, not a quality product. Here's some sample data. If anyone knows any more about this, I'd appreciate any input. The KID is set to 'PC' mode (just like it was with 1856)

(spaces added for readability)

1856
BatV,PanelV,PanelW,KWhrs,Ahr,Btemp,WBjrA,SOC,Ahr Remain, Status(?) [NEWLINE]
280 ,863      ,252      ,11       ,840,314    ,164    ,99    ,412             ,5,

1860
BatV?,PanelV?,?,?   ,Ahr Remain?, Status? [NO NEWLINE]
260    ,193       ,   ,12 ,440               ,2

[ClassicCrazy]

Below was the info from 1820 firmware . If the serial output is  different now with the new firmware and not documented you could open a ticket with Midnite and ask them .

Larry

-- PC Comm needs to be selected and saved
   - it sends data once a minute
   - sends data in raw decimal separated by a comma
   - data needs to be massaged to be read correctly and divided by 10
List of what it sends in order:
   * Displayed battery voltage
   * Displayed PV Voltage
   * Displayed output watts
   * KWH
   * Amp Hours
   * Battery Temperature
   * WBJR Current
   * Battery SOC
   * WBJR Amp Hr Remaining

Open a serial terminal and use the Baud Rate displayed in the screen when PC selected. PC comm only works with systems with a single unit.

[bmrc]

Thanks Larry. The 1856 manual shows the same info. I haven't found a 1860 manual, but the dataset is clearly different - It doesn't even have the same number of data points, and each sample is no longer followed by a newline. 

How do I go about starting a ticket with MidNite? This forum is the closest thing I've found to any sort of support. Clearly I didn't research this side of things well enough when making my charger selection  .

[bmrc]

I found it - http://www.midnitehelp.com/ . Thanks again Larry

[ClassicCrazy]

Yeah well stuff happens with new products and hard to figure out why something that works for so many can have a problem on specific systems.  Some companies don't even bother to air all their flaws out in public like on these forums and do firmware updates to fix stuff. So you have to give a lot of credit to Midnite . If you had gotten the Chinese controller and it did the same thing would you be able to get any support ?  I understand your frustration especially with a remote system like you have .  But maybe happy times are in the future for you !

Let us know if you find out what is going on with the serial data stream on this new firmware version. 

Larry

[jamesmc]

Any news on the serial stream?  Maybe the latest firmware (1860-1.1) fixed it?

[FNG]

I pinged Sue on this, Hopefully MidNite can respond

[KyleM]

There is a internal ticket being worked on for this issue.

If we find any information I will try and update this post for everyone else.

-Kyle

[bmrc]

Any news on the serial stream?  Maybe the latest firmware (1860-1.1) fixed it?

My understanding is that 1860-1.1 is the same as 1860, only with the new firmware flashing tool instead of manually updating with the FUP file. Can someone confirm?

[bmrc]

There is a internal ticket being worked on for this issue.

If we find any information I will try and update this post for everyone else.

-Kyle

As part of my ticket, Kyle requested some data logs. I sent the 2 logs attached to this post. Take note that the filenames start with the firmware version. The reply I received is...um..interesting;

This is the message that our Lead Engineer sent to me.



"It seems like the serial terminal is omitting the 0s and looks as if the spaces were empty, but the actual data is still the same format. here is a snapshot of the data that was sent:


130,565,,8,610,136,-21,100,821,15


13.0 * Displayed battery voltage
56.5 * Displayed PV Voltage
,, * Displayed output watts
.8 * KWH
610 * Amp Hours
13.6 * Battery Temperature
-2.1 * WBJR Current
100 * Battery SOC
821 * WBJR Amp Hr Remaining
15 * Reason For Resting


At this point in time the unit is not producing any power that is why the Watts is displayed as empty(0W)

138,832,229,1,70,135,152,97,798,5

This other snapshot looks like it is producing 229W
Another note to keep in mind is that KWH and AH will reset to 0 every night because those values refer to the total amount of power accumulated during the day."

There might be a setting in Putty that needs to be changed or configured differently to rreflect what you are used to seeing.

Please let me know if you have any further questions

[bmrc]

My reply was;

Please note that the logs I sent are from 1856 and the new 1860 firmware (version is in the filename). 1856 was sent only for reference, but thank you for clarification on what each data point is.

The issue is that the 1860 data does not have as many points. Here is a sample;  259,192,,12,440,2 (6 data points vs 10 previously).

Concerning putty - I've used the same 'saved session' (profile) to capture data from both versions. In 1856 I get one sample per line. In 1860 there is no line break. I'm only pointing this out to show that (besides missing data points) something did change.

It's been a week since I've heard anything else. I realize my situation may be somewhat unique and I can be a snarky bastard, but this is no excuse for no communication - especially when your product has already let me down once. My confidence in this brand is faltering.

[bmrc]

There might be a setting in Putty that needs to be changed or configured differently to rreflect what you are used to seeing.

Sit back for a second and realize the absurdity of this statement. You changed something in the firmware, you figure out what needs to be changed. I'm not poking around with settings blindly because 'there might be a setting...'. IT WORKED BEFORE YOUR UPDATE.

I'm an IT nerd, I work with data all day. If I changed say, the format of some XML data and then told a customer that 'there might be a setting that needs to be changed' but didn't provide details on what changed and why, I'd be out of a job.

Release the source code for the firmware and I'd be happy to figure out whats wrong on my own 

[KyleM]

The Forums are a great place to exchange ideas, and look for assistance from not only the Midnite Team, but other users as well.

We try and help everyone out as much as possible, but sometimes that takes time.

If you are having troubles with a supported aspect of our product, and are looking for timely Support on that issue. We have a trouble ticket system in place for that.

http://www.midnitesolar.com/support.php

Please feel free to submit a ticket and our Support Team will do everything they can to help you.

Thanks,
Kyle

[bmrc]

  *SMH* Kyle... I already had a ticket - #KID11477 . You closed it despite my response that you analyzed the wrong log . I've posted the entirety of the exchange here so others can see the process (or lack there of).

My gripe is the lack of communication. I clearly responded that the issue was not resolved, yet this message was not added to the ticket and it was marked as resolved. Customer Service 101 - don't close tickets until the issue is confirmed resolved or the customer hasn't responded in a reasonable amount of time. Neither were the case here.

I'm really not trying to be a burden, but I'm not going to get walked on. Midnite is consistently dropping the ball here...

[bmrc]

Just received this via email, ticket is still marked as resolved.

I have sent your request to release the source code to our engineering Supervisor.

When I get an answer I will be sure to forward it on to you.