APL's Pressure Contact Battery box

I think the only way to know if all the cells are making good contact is to run a capacity test. If you discharge it until the first cell group hits cutoff, you can then measure all the cell group voltages at the BMS connections. If the measured capacity is close to expected, things are good. If you have a group with one or more cells not making contact, the voltage on that group will be lower than the others and the measured capacity will be lower than expected.
 
fechter said:
I think the only way to know if all the cells are making good contact is to run a capacity test. If you discharge it until the first cell group hits cutoff, you can then measure all the cell group voltages at the BMS connections. If the measured capacity is close to expected, things are good. If you have a group with one or more cells not making contact, the voltage on that group will be lower than the others and the measured capacity will be lower than expected.

If there is a way to see individual voltage up on charge cycle it will show strait away. The group of cells with a fault will be far ahead of the others in series
 
APL said:
Anyway, I need to figure out a way to drain the pack down a volt or so.
What about the tried and true old school electric cooktop coils?
 
Indeed, I usually use a resistive load, like a space heater. If you can watch the cell group voltages in real time via a smart BMS, any decent load or charging will show a noticeable deviation on a group who's capacity doesn't match the rest.
 
I sent off for a cheap $30. 72v sensor controller from the bay, so I'll be able to put some stress on the pack with that for the time being. Still need to put the halls in the motor, but I wanted to beef up the drive gear & oil things up in there anyway.
(I have some time to do that now that the pack is mostly done.)

I have a new Votol EM-100 'clone' controller that I plan to try this summer, 80A - 3000w. Looks beefy, but might be junk, and will probably need tweak time,.. so the one just I ordered will fill in for now.

The next test is to drain the cells down from 65v to 64v, open the box, and measure all the cells again. That should give an early indication of contacts I think.

Another thing I thought of was to use Mylar sheet between the contacts to isolate all the cells except for one series string at a time. Measure the volts on each string.., but I'd have to open -close and re-torque the box 11 times. Yikes!

Anyway, back to the drain for now,.. yea 65v isn't far from 110v appliance voltage, and I'm thinking heaters and coils too. I have an old Variac 110v transformer that weighs about 20 lbs that will take a lot of heat. I think those things are just a big variable resistor
right? Plug a space heater into it, run the pack through the Variac, and just dial up the watts? Could hook the CA up to it.

(Can't do any charging yet because everything's being shipped, and still need to splice the new BMS connectors in.)
 
A variac is really a transformer and needs AC to work. Any old heating element will work. I've used a hair dryer that had a small DC motor (many do). I fried the power switch on the hair dryer trying to turn it off and it arced inside. I learned I can use the switch to turn it on but need to unplug the plug to turn it off.
 
That's the trouble,.., most of the AC stuff has a diode or a board or something. Even the space heater has an AC motor.
How about light bulbs in series? They're safe and I have a 6 bulb strip already. Would 6 x 100w-bulbs at 65v draw around 300w?
 
You can drain some small amps (0.5a-3a) from single cells (~3-4.20v )is pretty easy with toasters using about 1-5" of the 35" wire coiled around the backing board. Others used stove top and oven coils, hair dryers, maybe even clothing irons.

I was trying to find the post where a person bought like 5-10 space heaters I thought it might have been doctorbass or another :es: guru because you know how they love high voltage with lots of discharge current.
 
Some discharging threads, sorry not all directly relevant to this use case. A small fraction of what I've collected...

https://www.rcgroups.com/forums/showthread.php?p=48054827

reco mrforsyth
https://youtu.be/htY2mHTx3kQ

https://hackaday.com/2020/02/11/build-your-own-active-load

https://hackaday.com/2020/02/28/a-simple-yet-feature-packed-programmable-dc-load/

https://hackaday.com/2017/02/28/beefy-100-amp-electronic-load-uses-two-mosfets/
https://hackaday.com/2013/10/28/building-a-dc-constant-currentpower-electric-load

https://www.rcgroups.com/forums/showthread.php?1107072-How-to-build-a-constant-current-discharger

https://www.rcgroups.com/forums/showthread.php?3189173-New-LiPo-discharger-Maylibet

https://endless-sphere.com/forums/viewtopic.php?p=1625753#p1625753

https://www.rcgroups.com/forums/showthread.php?1255751-Poor-Man%C2%92s-5-Regenerative-Discharger

https://www.rcgroups.com/forums/showthread.php?p=48077255

https://www.rcgroups.com/forums/showthread.php?p=47765989

https://endless-sphere.com/forums/viewtopic.php?p=1592917#p1592917

https://endless-sphere.com/forums/viewtopic.php?p=1621117#p1621117

https://www.rcgroups.com/forums/showthread.php?p=42697527

https://www.rcgroups.com/forums/showthread.php?3490399-High-Power-LiPo-Discharger-Project

https://endless-sphere.com/forums/viewtopic.php?t=113163


 
I just use travel hair dryers. Set them at 110v and they work great. I even have a parallel outlet to plug in multiples for load testing.

20220127-114205.jpg
 
Wow, thanks for the help guys, there's almost enough info for a sticky thread already, especially with john61ct's personal archives! :)

I should build a nice Fet-driven variable drain, but don't want to get caught up in yet another project when I'm trying to get back to the bike build. So much to do yet on the bike, and I'm real happy to say that spring is coming up fast!

Lots to think about. I'll go into town today and check out the local market, but fully expect to be disappointed as usual. I think all I need for now is a 300 -500w drain to drop the pack volts a bit, and I'll use the controller to ride-test for amps once I get it hooked up,.. should be good for 1500w, and later on I can try the 3kw controller on it.

The hair dryer set up looks good. I don't understand how the motor can work on AC and DC, but if it jerks it works I guess. :thumb:
 
The hair dryer I use has a small DC motor (12-24v?) and has a full wave bridge rectifier ahead of it. It runs off a tap on the heating element.
 
I bought a cheap $10. hair dryer at Walmart, since I don't have one, and it worked like a charm! Thanks for the suggestions. :thumb:
I was a little worried when I saw the ground fault buttons on the plug, but it didn't matter.

Drain test.jpg

Made a little three way adapter and drained the pack down to 62v in about half an hour. Opened up the box and measured the cells,.. had to measure two in series along the strips, but they ALL came out to 7.83 - 7.84v, which is 3.92v each!

Cells all 3.92v.jpg

Jump'n Gigawatts, it works! :shock: How did that happen? Here I was all ready for a long battle, but every cell is making contact! :)
 
Yes, for sure, I have to thank everyone who helped me with this project, plus all the people who contributed to similar projects
that I researched here on the ES! :es:

Thanks ES members! :bigthumb: I could never manage to get any of this stuff done on my own.

What I meant was more like "how was I that lucky,. having 176 batteries all making contact the first time out is very lucky.

Not out of the woods yet though, I still have to draw some serious power out of it and run it through some real world rigors & check the temperatures. Some of that can be done soon, and some will take time, but at least I'm over the hump for now.

I forgot to weigh the box without cells, but I weighed the complete battery yesterday, and it came out to about 28 lbs.
I'll be taking it apart many times, so I'll get the box-only weight soon.
 
Not much to report yet, still trying to get the bike together enough to do a 1.5Kw test on the pack,.. not all that much but a start.
Lots to do yet, and more to come. :wink:

The silicone sheet 'spring' seems to be working, but I don't know how tight to turn the screws. Once it flattens out I can give them 1 - 2 more turns and it seems like a lot. What I need is a few dummy cells that have digital scale sensors inside to tell me what the load is. Otherwise, it's just trial and error.

If I build another pressure box like this, some improvements would be to 3D print the mid-section & sides, and with more clearance on the top and bottom for 'angle'-copper bus strips to pass more current, and provide room for more B+ & B- output wires.

Presure plate cooling.png

Pulling all the current from just one side of the bus is not the best arrangement, unless the bus copper is plenty thick enough.
I think I'll be alright with the loads I'll be using most of the time on this pack.

Another idea, just incase internal pack heat ever became an issue, is to use an aluminum plate for the bus-side, with combs or fingers to draw heat out between the cells. (thru-bolt holes) Just a thought.
 
APL said:
What I need is a few dummy cells that have digital scale sensors inside to tell me what the load is. Otherwise, it's just trial and error.

For this i have used FSR402 sensor with arduino.
 
Thanks agniusm, that's just what I needed,.. I looked around a bit and found lots of sources and info on it. Looks perfect! :thumb:
The sensor itself is only .5mm thick.

https://wiki.seeedstudio.com/Grove-Round_Force_Sensor_FSR402/

Grove Seeeduino.jpg

I was thinking that instead of a battery based sensor, it would be better to place the sensor in-between the cells at a thru-bolt location, and have a permanent pack pressure indicator. Check it over time and tighten as/if needed.

One more thing to develop and build to the list. (Not sure how to calibrate it, but I'm sure that it's all in the screen settings.)
 
Having trouble moving the bike build along enough to give this project decent test results. BMS connectors, motor tear down, wheel build, controllers to fool with, and chain line issues are all giving me flack. Not exactly good riding weather yet either.

Guess I'll move back over to the bike build section a little more and show how the build is coming along there. When it's back on the road I can continue to post the battery pack results here. I have the current tests, and some dielectric tests to show yet. Other than that, the pack works,.. just don't know how well.

I'll post a link in case anybody's interested,.. but for now there's not a lot I can do to stress this pack out enough and get any decent info, needs a bike.
 
You could theoretically calibrate using a weight. Having a weight constant you could calculate kilogram force or newtons per given area
 
Congratulations! :) Fingers crossed on the testing but not many could make an electro-brick look so damn cool ;) Grinding the cell ends could work out well, there's bound to be some variation in height so if it passes with slightly uneven reclaimed cells that's some added confidence in the clamping method. That "rub on silver plating crap" might be worth considering, I've never had all that much confidence in it either but the same chemical plating process is often used with industrial gold and silver plating.
 
agniusm said:
You could theoretically calibrate

Yep, close enough anyway, it wouldn't need to be perfect, just need to know how many lbs the cells are getting so they don't get damaged, and if the spring sheet is losing any spring pressure over time. I'll keep thinking about it. I wonder if I could hack one of those digital micro-scales and switch out the sensor with the FSR402. (basically just a resistance value, 100k - 200k)

Thanks Stan! Yea I saw the rub on silver plating stuff, looks a little iffy. I'll be interested in seeing how long it takes for this copper to get ugly at the contacts just the way it is. I just need it to make it through the season. If it's sooner than later, then maybe the paste will do it, and if not, then the plating is needed. I would actually prefer the raw copper because it's easy to keep clean with just a scuff pad.

Well who knows,.. gonna find out though. :)
 
OK, more APL craziness,.. as I keep playing with the threaded cell holder idea, maybe it's not viable, but I haven't seen threaded yet.

There are 'some' advantages, such as sealed caps and cans, and maybe 3D printable, but still a lot of problems to overcome yet.
Strength is all important of course, and a single strip wouldn't have that much, but there could be a socket plate that fits on the neg ends that would put the cells between two plates.

Since it's difficult to screw in cells that are packed tightly together, I thought maybe just series strips of 13S - 20S, or even shorter sections. 'O' rings could provide the spring for cell compression.

Anyway, just getting started on the copper links and how they 'might' look. Looking at it makes me think that maybe there could be sockets on both sides of the strip, and could seal the copper and double the P count.

Hmm.. well, if there's any interest then suggestions are certainly welcome, and it's just a whimsical idea that's probably doomed from the start. :wink:

Threaded cell holder.png

Series links.png
 
You're getting more than just a little good with CAD, impressive stuff! Idk for the spring, really progressive seems like the ideal, same pressure over the full travel. Balance bars? Cooling's always been a consideration here too, last attempt was using wedges to hold it all together (the dark ages sometimes works with 3d printing) and it left a great airflow, a zig zag pattern over the cells.
 
Thanks Stan, I still have a lot of troubles with CAD, but manage to get a screen shot at least. When it comes to organic 'blob' shapes I'm at a complete loss.

Do you know if threads can be printed with 3d very well? Not a lot of experience with 3D prints yet.
 
Back
Top