Has anyone used wire as a fuse with the Vruzend kit instead of the busbars for parallel connections?

mysurveymail

10 mW
Joined
Apr 17, 2019
Messages
20
Hello y'all,

This is my first real question to this forum. So I am asking for a push forward.

My application is upgrading my 36v 750w brush-less hub motor SLA powered EZ-3 trike.
View attachment 1


I have been using 3s3p SLA batteries for the last few years. During that time I have been scavenging 18650 batteries (200+) to "upgrade" my range, weight, and performance. So I bought a couple of the Vruzend.com 1.5 kits. I have been discharging and charging my 18650's with a couple of smart (4 slot) Zanflare chargers.
20190407_150654 (Small).jpg
Please note that the above photo is meant to show all the typical readouts for each cell IN THE PROCESS OF BEING CHARGED. These cells happen to be in the normal state of charging at 500mA not the "NOR" or "fast test" function, BTW.

Diligently using a "Sharpie" marker to write the mAh, mR, and voltage onto each cell discharged and then charged.

My proposed battery will be 10s10p, (or 10s20p if the gurus here advise) using the vruzend.com 1.5 kit.
stopmotionvruzendcolors-33-150x150.jpg
stopmotionvruzendcolors-35-150x150.jpg
My concern is, "cell matching, and parallel fusing."

So I am interested in using fuses or preferably some thin gauged wire as fuses in my forthcoming parallel pack. Mostly because my cells, however very thoroughly tested, are cells that have been recovered from disparate laptop batteries. I have been reading a lot on endless sphere and other sites and wondering if the "littlefuse" strip type fuses could be used with my vruzend build? I am planning a 10s10p battery pack with cells as close in specification, voltage, mAh, and mResistance as possible. Your thoughts on this would be greatly appreciated. Better still vruzend should offer some fuses for people uncomfortable just using their (great) battery building kit.

Cheaper and maybe better still would be using thin gauged wire to fuse between the POSITIVE terminals of the PARALLEL batteries. On the negative side of the vruzend kit I plan on using their standard buss bars, as well as using their standard buss bars in the 10s series stacks. So after a long explanation I want to wind the 30awg? wire from positive terminals to the adjacent 10s battery. What gauge or what do y'all suggest?

I will NOT be using a BMS nor a balance charger. I plan to "bulk charge" to 41 volts at 5 amps (divided by 100 or 200 18650 cells). I will NOT lie down and be flamed by hysterical BMS fanatics. My 36vdc 750w motor draws a maximum of 20.8A (without regard to start up, nor the non-existent hills here in south Florida (flat, flat, flat)). THAT is also the reason why I want to fuse my low cost cells.

Any knowledge would be greatly appreciated. I don't want to get flamed or into a non-experienced based or non-scientific argument. Please help PUSH me FORWARD.

C
 
Has anyone used some AWG wire as a fuse with the Vruzend kit instead of the busbars for parallel connections?


I am considering this in a 10s10p battery. My 36 volt 750 watt hub motor draws ~20A at 36V. I have purchased solid core tinned copper 22AWG for this purpose.

My concern is that a faulty cell in one of the series strings my "take out" the whole 100 battery pack. So if I understand what I have read all over the e-bike forums is that the parallel connections would be a good place to "fuse" the whole pack.

Obviously putting a fuse at the end of each series would also be a prudent place to add a fuse, I think?
How does 36Vdc drawing 20 Amps nominal translate to a 12v automotive fuse?
I plan on using a 12v 30 amp fuse and fuse holder.
8878.jpg
Are 30 amps just 30 amps regardless of the voltage, since the amp draw (wattage) is still just 750 watts? And therefore are these automotive fuses and fuse holder okay to use in an e-bike battery build?

Any experience with a particular gauge and type of wire to use (in AWG please) to fuse the parallel "strings"? How and where they used this in a VRUZEND kit?

Thanks,

C
 
I am about to do some experiments with making fuses, but I don't have any data yet.

The two places I have seen the most data about using wire as fuses is "second life storage" forum, and also "AveRage Joe" channel on youtube.
 
I don't do fuses on small batteries. I would just have a set of sense wires for parallel cells checking as 5s. Even with a bms. These is so you have access to balance with a thru the sense wires rc balance charger . Get the right one. Just make sure the connections are tight and secure. Do you have a C.A. ? These will help you if a cell is going south. Yes you will fill went one cell is having trouble
by the seat of your pants. If you have a smart one.
The more in parallel the better.
 
999zip999 said:
I don't do fuses on small batteries. I would just have a set of sense wires for parallel cells checking as 5s. Even with a bms. These is so you have access to balance with a thru the sense wires rc balance charger . Get the right one. Just make sure the connections are tight and secure. Do you have a C.A. ? These will help you if a cell is going south. Yes you will fill went one cell is having trouble
by the seat of your pants. If you have a smart one.
The more in parallel the better.

Thanks "Spinningmagnets" the posts you've made here on ES have been grounded in science, insightful, and have given others some real meaningful answers.

I have seen the youtube videos you referred, which is why I brought my questions here to ES. My hope is that once I finish individually smart charging my 18650's I will be able to build a 10s10-15p battery using the vruzend kits I have.

Meanwhile, I will take some of the solid core tinned copper 22AWG wire I have and do some tests for carrying loads, and what it takes to use it as a fuse. I have enough sub-par batteries, enough 22AWG, and some resistive loads to get some data. Then I will post it here.
 
mysurveymail said:
Are 30 amps just 30 amps regardless of the voltage, since the amp draw (wattage) is still just 750 watts?
Watts are the product of Amps times Volts.

So 30A at 750W has to be 25V.

If a supply can only put out 750W, then as V rises A falls and vice versa

If a component can only handle 30A current, then voltage has to increase in order to allow higher power, or you need to accept lower power in order to lower the voltage.

So a "1200W 50A" supply can deliver full power at 24V, but at 5V is limited to 250W only.

Hope that helps


 
When I said make sure they have a good connection I was talking about the Vruzend kit as some people were having trouble with good cell and cap contact. With used cells you will have to closely monitor your pack and not rely on just a bms. And if one cell went bad your should find out in lose of capacity. How many ways will you be monitoring your pack ?
 
john61ct said:
mysurveymail said:
Are 30 amps just 30 amps regardless of the voltage, since the amp draw (wattage) is still just 750 watts?
Watts are the product of Amps times Volts.

So 30A at 750W has to be 25V.

If a supply can only put out 750W, then as V rises A falls and vice versa

If a component can only handle 30A current, then voltage has to increase in order to allow higher power, or you need to accept lower power in order to lower the voltage.

So a "1200W 50A" supply can deliver full power at 24V, but at 5V is limited to 250W only.

Hope that helps

Thank you for your reply.
Maybe I misspoke. My hub motor is 750W and my controller is 36V rated I am sure to at least 25A. My question is what gauge wire to use in my 10s10p 18650 for the parallel connections?
I would rather have a fuse wire instead of the bus bars that are included with the vruzend kit.
I purchased 22awg solid core copper wire to use as a fuse. From each series string (10) I plan on using an automotive fuse rated to 30A... since my 10s pack will generate 37v nominal, and my amp draw will be ~21A will a 30A automotive fuse accurately fail at 36v?
 
mysurveymail said:
Maybe I misspoke. My hub motor is 750W and my controller is 36V rated I am sure to at least 25A. My question is what gauge wire to use in my 10s10p 18650 for the parallel connections?
Measure the amps actually flowing at each circuit in question while at WOT.

> I would rather have a fuse wire instead of the bus bars that are included with the vruzend kit

AKA fusible links.

Again I would test IRL rather than rely on internet sourced advice.

A/B testing the fusible link idea against an appropriately rated circuit breaker (expensive) or proper fuses.

Sounds like a dangerous kludge to me, but I'm sure someone with solid education and more experience could safely guesstimate.

> will a 30A automotive fuse accurately fail at 36v?

Such fuses' behaviour varies widely between model / lines / makers, even between the top brands like Bussman, Eaton, Blue Sea.

Also changes as the CP device breaks down with usage over time.

The data-sheets should have charts for the different timing-ratings desired at various ampacities.

Do not use cheap generic components, industrially engineered or EV-specific.

Not at all a straightforward topic, only after a few hundred hours of study, and ideally IRL experimentation, will you realize how little you (and I) know about it.

Until then Dunning—Kruger rules.

KISS is a very important fundamental principle here. Don't forget to take failure modes into account.




 
This link tells us that 22G copper wire is 25 mils in diameter, or .0252 inch.
https://www.rapidtables.com/calc/wire/wire-gauge-chart.html

W.W. Preece calculated fusing currents for wires. This source, (skip the math if you wish) has a table of fusing currents for various diameter wires.
http://www.nessengr.com/technical-data/fuse-equations/

And Ness has an entry for something close to 22G. The table says 40A. No good for you.

Let's say you wanted copper wire to fuse at 6A. The closest entry is for 5A, and the Ness table says wire diameter is 6 mils. The other link says you're looking at 34G wire to get that thin.

You can calculate the resistance of 34G wire with this link.
https://www.cirris.com/learning-center/calculators/133-wire-resistance-calculator-table

You can see from the table why fuses are made out of lead-tin. I popped the fusible link in my car's cigar lighter running an air pump once. I could see it was some kind of lead wire that had melted. I put in a thin piece of solder, made sure it would melt with a 20A load and made do with that.
 
So in effect fusible links are a non-starter here,

when energy efficiency is desired, need the low resistance of much higher gauge wiring?
 
Thanks for the information y'all,

spinngmagnets , john61ct , and docw009 thanks.

I think after all this I will still not use the buss bars in parallel supplied with the vruzend kit. Unless gives me a compelling reason to do so. I am using tested, but recycled laptop cells of verying age and capacities. On repackr.com and better still some guy on YouTube "rinoasupergenius?" shows how he separates 18650's by capacity, then combines them in such a way that each series string has nearly the same capacity as measured in mAh. He seems to be using 30AWG tinned copper in his connections to his buss bars... kind of like other builders I have seen, and Tesla Motors. My fear using these recycled batteries is ONE cell shorting out my whole array and starting a fire. This is why I am interested in using thin gauged wire even in the 10s wound from positive cap to positive cap on the vruzend caps. Until I get some further guidance as to what actual AWG to use.

Until then I will just put a fuse at the end of each 10s battery.
View attachment 1

So each 10s would be electrically mechanically attached (using the vruzend caps and buss bars), but not electrically connected in parallel until after each series string's current passes through a 30A fuse.
resist-loosening-768x445.jpg

The 10 strings will then come together after the fuses. I hope this will give me the same Amp Hour gain because I will end up with 10 in parallel.

This is the way I have wired my existing 3s3p 12v SLA battery pack. Which if I did the math right each battery string of 3 batteries being 12v and 7aH gets me to 36v and 7aH per pack times the 3 packs should be 21aH... But my understanding is that SLA batteries don't deep cycle as well as 18650's and I am not getting the full amp hours out of my current configuration, plus I think they weigh more than my 10s10p pack of 18650's will.

Thanks again,
C
 
When playing roulette with use cells you should put a set of sense wires on so you could monitor and have access to the pack for self balancing and per cell charging or with rc style Charger good balance 5s at a time cheaply and monitor with a cell leg or something like. Knowing is knowledge.
 
999zip999 said:
When playing roulette with use cells you should put a set of sense wires on so you could monitor and have access to the pack for self balancing and per cell charging or with rc style Charger good balance 5s at a time cheaply and monitor with a cell lock or something like. Knowing is knowledge.

Thanks 999zip999,

I plan on using a bulk charger once my battery is built 100 batteries at 500mA per cell less if I goto a 15p battery, but only to 4vdc. Then after some bench testing with resistive loads (3 car headlights wired in series) and a shunt style amp meter, I will then test each batteries voltage again.
For the last month I have charging discharging and recharging my batteries with two of these smart chargers.
20190407_150654 (Small).jpg
Recording each cell with mAh, internal resistance, and final voltage. In the next few days I have done this to over 200 18650's. Then I begin assembly of my 10s10p or 10s15p battery.

Any info on fuses or wire gauge for parallel fusing?
 
I "finished" my first 10s10p battery using vruzend 1.5 kit

The longest part of all was that getting my cell count up to 100 used cells from laptops took forever. I started with over 200 18650 batteries. I used 2 4 slot "zanflare" chargers to do a "fast test" on all my cells. This meant a discharge first then to charge. This was so I could record the mAh capacity, internal resistance, and their final voltage.

I started out dividing them by capacity, then trying to make each series of 10 add up to the same overall mAh. It was somewhat maddening, so I gave up. Then I went to repackr.com and typed in all my cell capacities and let their website design all my 10s strings to be matched in Ah capacity. They managed to get the apparent capacities very very close. Cool. This seemed to be from everything I read to be the best method to get the maximum amp hours from my battery.

I started assembling my vruzend caps based on Micah's video on YouTube. That was fine until I realized that since I was starting out with a bunch of differing capacity cells (not all the same bought new). So based on repackr's design I had to build my caps in a SERIES fashion. This meant alternating the vruzend caps, red, blue, red, blue, etcetera. Then I had to pay attention to which direction the "dovetails" on the caps were oriented, so that I could successfully join each series string to the parallel caps in the array. It took a while, but went well. I have not yet put on the charge and load connectors yet, but that is quite easily done.

One suggestion that I do have which would have made putting the 200 nuts on would have been a magnetic nut driver. The nuts with this kit are nonmagnetic, so each one should be started by hand, and finish torqued with vruzend's nut driver.


My next challenge will be to bench test this battery with a load (not my ebike). I want to see how the cells hold up to a similar load as my 36v 750w brushless hub motor. I have thought to run a few 12v lights in series, much like Micah did in one of his YouTube videos. My load would ideally be 20 amps, so my small 12v garden lights are not adequate. I don't really want to buy some big resistors and just generate heat. I will be researching the wattage of automobile headlights to see if this is a viable option. Any suggestions would be greatly appreciated.

After half of my battery was built I realized that it would be better to "split it in half" to make it fit better in my battery box. So the photos attached reflect this. When finally wired onto my bike it will be the same as 10s10p (although physically each pack will be 10s5p). The weight of this pack is 16lbs, compared to the 36lbs of 9 (7Ah) SLA batteries I have been using for the last 2 years.
View attachment 1
20190501_184607 (Large).jpg
 
Back
Top