diy solar

diy solar

How do i capacity test with no equipment?

Most shunts are pretty accurate, most clamp meters are ballpark. That's my experience with the cheaper ones, anyway.

Shunts can run hot when run near their capacity, since they are resistance based. Get a bigger shunt if that is a problem, generally speaking you want to run about midrange of the shunt rating. Shunts are quick and easy to calibrate in the factory, so accuracy is rarely a problem in bundled units.
I believe that. I don't have a clamp ammeter so haven't reviewed their accuracy. But, it basically boils down to whether you're going to buy something calibrated like a good quality shunt monitor (and therefore rely on the calibration that the manufacturer did), or rely on something else such as a clamp ammeter or your BMS's integrated shunt, and rely on yourself to calibrate it. I don't think an $18 board from Alibaba is likely to have a really well calibrated measurement, but who knows, maybe they do calibrate each board and it's actually very accurate.
 
I believe that. I don't have a clamp ammeter so haven't reviewed their accuracy. But, it basically boils down to whether you're going to buy something calibrated like a good quality shunt monitor (and therefore rely on the calibration that the manufacturer did), or rely on something else such as a clamp ammeter or your BMS's integrated shunt, and rely on yourself to calibrate it. I don't think an $18 board from Alibaba is likely to have a really well calibrated measurement, but who knows, maybe they do calibrate each board and it's actually very accurate.
The actual process is cheap, fast, and doesn't require much technology or technical knowledge. I would rate as pretty unusual to find a shunt that wasn't calibrated. Takes about 60 seconds.
 
The actual process is cheap, fast, and doesn't require much technology or technical knowledge. I would rate as pretty unusual to find a shunt that wasn't calibrated. Takes about 60 seconds.
Sure, just it's a bit too expensive to do on all those Daly BMSs, right? I agree they probably do *something* in the factory, but how are you going to know if it's calibrated or not unless you also have a known-calibrated part to compare against?

Or, do they throw in a spec sheet with tolerance values for $18 plus $0.11 shipping? :). Just joking around obviously. I've seen some pervasive misunderstandings about shunt monitors in general on this forum so I wanted to try to help clarify some things.
 
Anyway, I'm not trying to discourage people from buying new fun electronics to support their hobby. Just giving an alternative above since you are probably only going to do a handful of capacity tests in your life unless you're running a YouTube channel or a battery/electronics company.

Here's a $0 method that might work well enough. It's going to be less accurate than a good shunt but maybe more accurate than a cheapo shunt:
1) find something which draws ~9A from your battery, according to your BMS (a couple light bulbs connected to the inverter output)
2) use the 10A mode of your cheap multimeter and install it in-line with the battery, just like you would that $18 board from Alibaba
3) measure 0A and 9A load, and fit a line to the multimeter-vs-BMS current readings. This gives you offset and gain correction.
4) remove the multimeter and thereafter just rely on the BMS's reading but correct it using the values calculated in the last step

To reduce the risk of step (2) blowing your multimeter fuse, you can apply a short wire across the ammeter leads until the current has stabilized for your measurement, then remove it.

As a bonus, now you can rely on your BMS's current readings for the lifetime of your battery, in all likelihood. Most of the error is in the shunt resistance variation which is fixed after manufacturing, not random voltage offset of the electronics, if they've chosen a low-offset analog front end in the BMS design.
 
Last edited by a moderator:
Anyway, I'm not trying to discourage people from buying new fun electronics to support their hobby. Just giving an alternative above since you are probably only going to do a handful of capacity tests in your life unless you're running a YouTube channel or a battery/electronics company.

Here's a $0 method that might work well enough. It's going to be less accurate than a good shunt but maybe more accurate than a cheapo shunt:
1) find something which draws ~9A from your battery, according to your BMS (a couple light bulbs connected to the inverter output)
2) use the 10A mode of your cheap multimeter and install it in-line with the battery, just like you would that $18 board from Alibaba
3) measure 0A and 9A load, and fit a line to the multimeter-vs-BMS current readings. This gives you offset and gain correction.
4) remove the multimeter and thereafter just rely on the BMS's reading but correct it using the values calculated in the last step

To reduce the risk of step (2) blowing your multimeter fuse, you can apply a short wire across the ammeter leads until the current has stabilized for your measurement, then remove it.

As a bonus, now you can rely on your BMS's current readings for the lifetime of your battery, in all likelihood. Most of the error is in the shunt resistance variation which is fixed after manufacturing, not random voltage offset of the electronics, if they've chosen a low-offset analog front end in the BMS design.
Two types of shunts, two methods of "calibration".
If you have a higher capacity shunt, notice that it seemingly has a notch cut out of one of the bars? That is the calibration, totally automated process, sends a current, measures how much out of tolerance, and adjusts the notch size accordingly.

Smaller shunts (like the $18 one) look like a half loop. Since the measurement circuit is built in (voltage measurement), the resistance of the shunt is once again measured by sending a current through the shunt, and an offset value is stored in non-volatile memory.

Nearly every component in any device is manufactured, measured, and or calibrated to one extent or another. Every manufacturing facility I have worked in, has their own calibration facility or team.

Some items measured in costs by fractions of a cent (for example resistors) have broader tolerances, and of course you pay more for tighter tolerances.

I'm not sure why you would think shunts are somehow special and just shipped hoping the customer won't notice. I can tell you that the 3 different (under $50) models of cell testers I have tried, all agree within a tenth of an amp hour with two different (complete with actual lab calibration certificates) models of lab equipment that I own.

I am sure there are vendors out there that sale terribly uncalibrated equipment, but they have a tendency to go out of business.

Daly seems to have their own method of business, and it will either adapt to meet the demands of the market, or wind up as a "remember this company" type of company. They do have a calibration method that they apparently don't want average users to perform. I have no clue why they don't do a better job in the factory, but it is hurting their business and reputation in a very competitive field.
 
Last edited:
I'm not sure why you would think shunts are somehow special and just shipped hoping the customer won't notice. I can tell you that the 3 different (under $50) models of cell testers I have tried, all agree within a tenth of an amp hour with two different (complete with actual lab calibration certificates) models of lab equipment that I own.
That's a great data point, so thanks in advance if you can share with everyone the model numbers of those shunts that you measured.

Why do I think that? Mainly because shunts integrated in BMSs are not calibrated very well, and because of the incredibly low price of some of these Alibaba shunts. I don't have actual test data like you do.
 
Last edited by a moderator:
Anyway, I don't want to get into an argument @Just John , I was trying to answer the main question of this thread:

How do i capacity test with no equipment?​

So, buying a shunt or battery tester isn't exactly "with no equipment". :)
I agree it would make it easier and work well!
 
Last edited by a moderator:
That's a great data point, so thanks in advance if you can share with everyone the model numbers of those shunts that you measured.





The cheap shunt based measurement devices I've used.

My Kunkin KP184 was calibrated on January 11, 2021, should I upload a copy of the certificate? It's in Chinese.
My (multiple) East Tester ET5410 also have calibration certificates, these two are what I compared results with.
Sorry, but yes, I've actually tested with calibrated instruments, and use common sense.

Why do I think that? Mainly because shunts integrated in BMSs are not calibrated very well, and because of the incredibly low price of some of these Alibaba shunts, mainly. I don't have actual test data like you do.

Nobody denies Daly does a poor job. Some other manufacturers do much better (so I've been told). JBD apparently does better, I'll let you know when mine arrive to test (they just cleared customs). There are users that report great results from calibrating their Daly, I haven't bothered to try.

Even the cheap testers above have calibration procedures, but I haven't bothered to try them (didn't see a need since they are pretty accurate).
 
should I upload a copy of the certificate? It's in Chinese.
Nope, I don't think anyone (including me) is doubting you or attacking you. Just asking for information/advice. Seems like you're upset.
 
Can you link to or explain how to calibrate the Daly? I've done some digging and cant find any info on it.
I would have to search on this forum. It was someone here on this forum, and he gave details about it, and how to set some things. Specifically, there is a "noise" setting that helps with the jittery results. It isn't called noise, but from memory, that was pretty much it's function.

I know you must do it from the Windows software, not the bluetooth app.
 
I would have to search on this forum. It was someone here on this forum, and he gave details about it, and how to set some things. Specifically, there is a "noise" setting that helps with the jittery results. It isn't called noise, but from memory, that was pretty much it's function.

I know you must do it from the Windows software, not the bluetooth app.
Thanks for that. I'll have to break out the laptop next weekend and go digging in the software.
 
Thanks for that. I'll have to break out the laptop next weekend and go digging in the software.
Yes, I haven't been playing with mine since it is out in the garage. I just set up a headless (dummy HDMI plug) Intel NUC, so I will probably start playing with it soon.

I'm no expert, I just plug things in and try to figure them out, same as you.
 
Thanks for that. I'll have to break out the laptop next weekend and go digging in the software.

I've checked and these are the same software and drivers Daly makes available.
It's just easier to find here.

There is a lot here in the resources section.

 
Most shunts are pretty accurate, most clamp meters are ballpark. That's my experience with the cheaper ones, anyway.
I bought my own clamp ammeter this week to debug some low panel production numbers. I did a quick 6A measurement by paralleling two of my Rigol DP832A outputs in constant-current mode to give 6A of DC current. Here's the measurement data measuring the same current with 3 different pieces of equipment:

Rigol DP832A (~$672) CH1+CH2: 6.002A
Innova 3220 Multimeter (~$25): 5.99A on 10A range
Uni-T B4Q094 UT210E Clamp Ammeter (~$50): 6.01A on 20A range

So based on this, the spread is about 0.3% of 6A, or if we think the Rigol is the most accurate it's about 0.13%. Whether that is an offset error (which would diminish in proportion to a larger current) or a gain error (which would show the same percentage at higher currents) is not known from this measurement. There is some offset due to earth's magnetic field, but I also zeroed the ammeter before using it, like you're supposed to do for DC measurements. To be on the pessimistic side, let's assume it's a gain error, then your 280Ah battery capacity would have a measurement uncertainty of about 0.13% of 280Ah or 0.37Ah. I think that's well within an acceptable measurement accuracy for a capacity test, but that's a matter of opinion.

Obviously, I haven't done a real capacity test though; I just got it in the mail today and don't have another piece of better equipment to compare against. But, based on this, I would just trust the clamp ammeter.
 
Last edited by a moderator:
Got an email from FastTech canceling the order for the capacity tester. No reason as to why. Guess I'll keep looking...
 
Got an email from FastTech canceling the order for the capacity tester. No reason as to why. Guess I'll keep looking...
I bought this one on Aliexpress: https://www.aliexpress.com/item/1005001458325206.html?spm=a2g0s.9042311.0.0.6dd04c4dYFsgNl

I liked it because there is both PC/laptop software and a BT app (Android or iOS) to connect and track the discharge. Of course, there is a nice color screen on it too, so you don't actually have to use either the laptop or phone interface.

It took about 6 weeks to deliver tho.
 
180 watt tester would take 3 1/2 days to test my battery, and thats at 100% load which nobody suggests with China components

I would like to do a .2C rate test and be done in a reasonable amount of time lol
 
180 watt tester would take 3 1/2 days to test my battery, and thats at 100% load which nobody suggests with China components

I would like to do a .2C rate test and be done in a reasonable amount of time lol
Depending on how you wish to test, easiest is to use a shunt and an inverter. I have been using multiple ET5410 units, but that is a significant investment. Good for 28 amps each on 12v packs.
 
Back
Top