diy solar

diy solar

Juntek 200A smart shunt counts negative amp-hours after reaching 100%SOC

RGIvy

New Member
Joined
Dec 9, 2020
Messages
85
I have both the Victron Smartshunt as well as several Juntek smart shunts. The Victron is some cool kit, very accurate and works well. At a fraction of the price, the Junteks were great too. No Bluetooth app but you could mount the wireless display anywhere in your van where it would make most sense.

Previous Juntek shunts would reduce the AH drawn to zero when you recharge the battery. After reading zero they'd stay there at zero, still showing a negative amperage flowing through to the battery (not really a physical possibility but you know what I mean). But they wouldn't continue counting AH drawn well into negative figures.

The latest batch of Juntek shunts do count into negative AH drawn. Each cycle this negative gets worse.
The problem is that when you once again discharge, the shunt first needs to get back to zero before it displays a reducing SOC. This just doesn't make sense and I can't see how this of use to anyone. Eventually I guess it would show you have 100%SOC when the battery is depleted.

Juntek - end after recharge.png

The display on the Left (Juntek) is "incorrect", taken from a new batch we just bought. The one on the Right (Junctek) is from an older batch and what I want to see.

I have contacted who I think is the manufacturer but it's taking loads of to-and-fro to get nowhere. The Alibaba supplier that we bought them through is probably contacting the same bunch but it's all taking a long time.

Has anyone perhaps had this issue and solved it?
 
This just doesn't make sense and I can't see how this of use to anyone. Eventually I guess it would show you have 100%SOC when the battery is depleted.

Maybe I am misunderstanding, but I think it would ultimately be more accurate. Due to battery inefficiencies, if you can't just take 100AH out and put 100 back in and say the battery is fully charged. So I would think having it go into negative territory would cause it to have to put more back in before it got back up to 100%. But I am not sure what that one reading actually is saying.

Also, is there not a way to "zero" the meter when you actually are at 100%?
 
Maybe I am misunderstanding, but I think it would ultimately be more accurate. Due to battery inefficiencies, if you can't just take 100AH out and put 100 back in and say the battery is fully charged. So I would think having it go into negative territory would cause it to have to put more back in before it got back up to 100%. But I am not sure what that one reading actually is saying.

Also, is there not a way to "zero" the meter when you actually are at 100%?
That would mean you have to zero it after each cycle, otherwise it would keep going further and further out of sync.
I find with all of these shunts after a full cycle (start at 100% SOC, discharge, charge back to 100% SOC) the shunt always seems to show that you put a little more back in than you took out. Even the Victron.

For example, you have a 100ah battery:
You start at 100% SOC.
You discharge 100ah. The shunt shows 0% SOC and shows that you have drawn 100ah from the battery.
You then charge back up to 100% SOC. It's gonna show you put in a bit more than 100ah (let's say 103ah for argument sake). So it's showing that you have 100% SOC and have drawn -3.00ah.
You then discharge again. It counts up the AH drawn until it reaches zero. All this time the shunt shows you have 100%SOC.
It reaches zero AH drawn and begins to show a positive AH drawn, and the SOC drops from 100% SOC.
You continue to discharge until the battery is depleted and the total AH drawn now shows 97AH and 0% SOC
You charge up again to 100% SOC and the shunt shows that you put in 103AH. That brings it to -6AH drawn, 100% SOC
..and it keeps going until the shunt reaches a point where it shows you have 100% SOC, 0AH drawn but in fact the battery is depleted.

.... unless you zero it manually every time it reaches 100% SOC
 
You charge up again to 100% SOC and the shunt shows that you put in 103AH. That brings it to -6AH drawn, 100% SOC
But would that not still be accurate for the SOC since you need that extra 6AHs to be fully charged? I would think that as long as the calculation for 100% is taking that "extra" into account, that is all that really matters.

In any case I was getting ready to order one of these, so I guess I can play around with it and see.
 
I would think that as long as the calculation for 100% is taking that "extra" into account, that is all that really matters.
The thing is that it's not.

When I last tested a full cycle on 240Ah cells, it had reached -7.5Ah drawn. That's around 3% of the SOC
When I began discharging it counted the AH drawn until it reached zero. At that time the SOC display showed 100% when in fact we'd drawn 3% of the batteries capacity so it should have shown 97%.
Only after we reached zero and began a positive amount of Ah drawn did the 100% SOC indicator change and begin to reduce. So essentially every full cycle my %SOC indicator will be another 3% out of kilter.
 
Every shunt based battery monitor I've ever used drifted in the same way. It's not a perfect calculation. Most had some way to fine tune the calculation, but they all had some way(s) to reset the capacity to 100. Some reset to 100% at float. Some you can reset when the BMS is in balancing mode. Is there a setting to reset the capacity?
 
Is there a setting to reset the capacity?
It is possible to manually reset, but why should we have to?

In previous models of the exact same shunt it would count DOWN the AH drawn until it reached zero. Then it would stop. The zero was reached when the battery was very close to 100% SOC. The little bit of "extra" charge that went into the battery was effectively ignored. This was good because it acted as an automatic "reset" to 100% SOC and 00.00AH drawn. At the end of each cycle the shunt would effectively have reset itself.

In the current model it doesn't stop at zero. Each cycle it goes further and further into negative unless you manually reset.

I was hoping someone would have faced this problem and solved it.
We've requested a full refund from the seller as neither the seller nor the manufacturer can help us.
 
Does the below make sense?
It's based on the widely accepted "fact" that you always put more back into the battery than you took out to get back to 100% SOC.

Screenshot 2021-09-24 at 09.43.33.png
Unless you manually reset each cycle it will become worse and worse. In this case a cumulative 2.5% error.
After cycle 40 the shunt will start with -280ah drawn and show a 100% SOC. Discharge the battery, drawing out 280ah, and the shunt will show 0ah drawn and still 100% SOC (whereas the battery is in fact totally discharged).
 
FWIW, the one I ordered is working fine, and has been tracking the same SOC as a different model shunt I also have connected. The only issue I have is what I consider to be a backwards way of measuring current flow. Every other meter I have used shows negative current flow when discharging the battery and positive when charging. This one is opposite.



On the one you are having issues with, do you have any alarm values set? Just curious if changing some other value from the defaults causes it to not work properly.

I just ordered a second unit because I mistakenly ordered the 100A one, and wanted the 300A version. Guess I will see how that one works when it comes in. Are both of yours the exact same model?
 
Last edited:
FWIW, the one I ordered is working fine, and has been tracking the same SOC as a different model shunt I also have connected.
Can you see a software version anywhere?
I'll check when I'm back at my test rig. I'm wondering if the 100A you bought is older stock before they introduced the change/error.

It's their way of doing it Guess, but I prefer the Victron shunt which shows minus out of the battery and plus into the battery.

On the one you are having issues with, do you have any alarm values set? Just curious if changing some other value from the defaults causes it to not work properly.

I just ordered a second unit because I mistakenly ordered the 100A one, and wanted the 300A version. Guess I will see how that one works when it comes in. Are both of yours the exact same model?
I don't know. I have put a 50A (VAT1050) side-by-side with a 200A (VAT1200). The 50A is fine. But I can't see any setting that I can change (been through all the menu options including hidden menus).

I'll find the software versions. The seller did at one stage mention that a change was made in the latest version to accommodate overcharging. Their words not mine.
 
The seller did at one stage mention that a change was made in the latest version to accommodate overcharging

Is it possible the one that is causing you issues is actually the older stock and the other one is the newer version? From that description, it sounds like they were fixing the issue you have with the one.
 
Is it possible the one that is causing you issues is actually the older stock and the other one is the newer version? From that description, it sounds like they were fixing the issue you have with the one.
We bought two identical shunts from the same supplier: one in March and the other in September. The earlier purchase is fine, the latest not. Maybe they sold from new stock in March and from old stock in September - but not likely.
 
We bought two identical shunts from the same supplier: one in March and the other in September. The earlier purchase is fine, the latest not. Maybe they sold from new stock in March and from old stock in September - but not likely.

If new stock was put on the top of the pile, the older stuff could be sitting around for a long time, waiting to be uncovered. I have seen it with other things (getting an older version long after a new version was released).

FWIW, the first one I bought, I got on ebay (still shipped from China). The one I just ordered is coming from Junce store on aliexpress. Guess we shall see when it gets here.
 
Last edited:
FWIW, the first one I bought, I got on ebay (still shipped from China). The one I just ordered is coming from Junce store on aliexpress. Guess we shall see when it gets here.

So I got the one from aliexpress and I can confirm the same "issue" as the OPer. What I will do is run them side by side for a few days to see if I run into the same discrepancy issue that was noted.

For me, it will not be a huge issue. I bought the second one because I actually needed a 300 amp shunt vice the 100 amp one I originally bought. The shunt/transmitter is interchangeable and it is the display unit that cause the problem. I can use the second one (if it actually is inaccurate) as just a one way current meter for another project. I don't need capacity for that.

<edit> If I plug it into a USB port on my laptop, it sees it as a "device". It would be nice if there was a way to update the firmware via USB.
 
Last edited:
Ok, so it only took one day to confirm that indeed this meter is not working properly. I had zeroed them both yesterday afternoon (when the original was already at 100%). I still had a couple of amps flowing into the batteries until the sun went down. This put the new meter at about -29ah by night time. Since my loads turn on during the night, this morning I went to check. The original meter was at 88% which is correct. The new meter was at 90%. So I agree that after a couple of days the reading will be totally useless.

BTW, I know the original one works correctly because it tracks identical to a different model shunt I was using. I had bought the Juntek due to it's nicer color display.

@RGIvy Did you ever get any resolution from your seller? I sent a message to mine, but have not received a reply yet.
 
That’s what I was hoping (in vain) for.

It may take a while, but if this is a result of a newer FW version, then perhaps after a while they will do something to fix it once people start complaining more. But I wonder how many people would even notice it.

I suggest people interested in this meter buy from an e-bay seller instead. It may be a few bucks more, but at least there is a slightly better chance of getting a response/resolution from the seller if you get one of the "broken" ones.
 
The shunt/transmitter is interchangeable

I have to walk this statement back.

Yesterday when I got the new shunt, I put it in series with the original (I removed the different brand one I was using for comparisons to the Juntek). On the original Juntek, I was using a USB cable vice wireless. I unplugged the USB cable from the original shunt and plugged it into the the new one. I then looked at the original meter and could have sworn it was displaying the correct voltage and current. But then I plugged the cable back into the original so I could monitor both overnight (the new one was connected wireless).

This morning I removed the original shunt and plugged the USB cable into the new one. Remember, I wanted to use a 300 amp shunt instead of just a 100 amp one. So after I finish, I go into the shed and see there are no volts or current displayed. WTF?? Well, after some trial and error I discovered that they are not interchangeable after all, even when connected via USB. Oh well.

I also noticed the 100 amp version has a wireless address in the A01-Axx range, and the 300 amp version has an address in the K01-Kxx range. Since I can not seem to change the letter, I don't think I can get them to talk to each other.

I am using the "broken" meter to monitor the output of my charge controller so I can see exactly how many amps it is putting out at any time (it has no display of it's own). Previously I was having to do some mental math to determine the charge controller output when I would have a load on the batteries.

Because I dislike that negative number the Juntek uses when current is going into the battery, I connected the new shunt in reverse. That solved that problem and now I see a positive number coming out of the charge controller. I set the battery capacity to 0 on this meter which causes the battery display to always show 100%. I'm not using this one for capacity anyway, so the battery % is meaningless. But it just looks ascetically better with a full battery graphic.

20211010_133208.jpg
 
Last edited:
Back
Top