The bistander thread

[bistander]

Not according to what I have read, but you are entitled to your opinion just as I am entitled to mine. I appreciate the warning though. You can never be too careful when dealing with batteries.

So where did you read that?
bi

[Turion]

" there are no negative effects", you say.

Have you considered the resistive loss? You know, I^2*R? So loss (& heat) proportional to square of current while charge is proportional to current. For equal charges, constant current vs pulsed, simple math proves there is negative effect.

Not according to what I have read, but you are entitled to your opinion just as I am entitled to mine. I appreciate the warning though. You can never be too careful when dealing with batteries.

[bistander]

I do it because trying new things is part of, you know, a thing called research. I seemingly suggest it to others because I have seen it’s benefits with lead acid, as far as the research goes there are no negative effects, and the verdict is still out on whether or not there are positive effects, and NONE of the current research was running the loads between the positives ANYWAY, so I have more experience with that than the “experts” do. The more people that TRY something, tge more information we get.

" there are no negative effects", you say.

Have you considered the resistive loss? You know, I^2*R? So loss (& heat) proportional to square of current while charge is proportional to current. For equal charges, constant current vs pulsed, simple math proves there is negative effect.

I ask these things, like why you do this or where did that come from, for the learning experience. That can go either way. Maybe I find out some are exploring Li pulse charging since I got out of the business, or you might learn it is a bad idea.

Another thing:
Charging is the hazardous half of the Lithium cell equation. You can ruin the Li cell with discharge abuse, but an accident on charging can burn down your house, injure or kill somebody. I know of several homes lost. It happens.Can your BMS tolerate pulses? I've tested many cells. From milliamps to kiloamperes. Used fireproof containers often. Lost some cells, mitigated some messes, but never hurt anybody or melted a structure. For smaller cells/batteries look into a "lipo" bag. Most test labs have a steel drum half full of sand under the test bench.

Personally I'd recommend to avoid pulse charging Lithium cells. And use a good functioning BMS capable of disconnecting the charge source. Be careful.
bi

[Turion]

I do it because trying new things is part of, you know, a thing called research. I seemingly suggest it to others because I have seen it’s benefits with lead acid, as far as the research goes there are no negative effects, and the verdict is still out on whether or not there are positive effects, and NONE of the current research was running the loads between the positives ANYWAY, so I have more experience with that than the “experts” do. The more people that TRY something, the more information we get.

[bistander]

I have experimented with pulse charging li batteries for quite some time. I don’t know that it’s any more effective than any other method of charging them and don’t claim that it is. But it works, so running loads between the positives with li batteries is better than using lead acid simply because the batteries have lower impedance. I made no claim about it other than it can be done.

Why do you do it and, seemingly, suggest it to others?
bi

[Turion]

I have experimented with pulse charging li batteries for quite some time. I don’t know that it’s any more effective than any other method of charging them and don’t claim that it is. But it works, so running loads between the positives with li batteries is better than using lead acid simply because the batteries have lower impedance. I made no claim about it other than it can be done.

[bistander]

Google it. It’s called research. Here is one article that came up immediately.

https://iopscience.iop.org/article/1...6/817/1/012008

OK. After reading a few and spot checking a number of hits, most appear to be related to fast charging, not what you're doing, correct me if mistaken.
______
https://www.researchgate.net/publica...ng_Parameters_ on_the_Life_Cycle_of_Lithium-Ion_Polymer_Batteries

This article did find benefit, of about 100 cycles ( out of 4000?) life cycle.
_____

This thesis found NO benefit. See quote below.


"However, with some confidence it can be claimed
that after these three types of testing, pulsing with a constant current magnitude and
duration does not produce any positive results, which is in agreement with some of
the reviewed literature."

from:.
THE EFFECTS OF PULSED CHARGING ON LITHIUM
ION BATTERIES
A Thesis
Presented to
The Academic Faculty
by
Daniel William Gaddes
In Partial Fulfillment
of the Requirements for the Degree
Masters of Science in the
George W Woodruff School of Mechanical Engineering
Georgia Institute of Technology
December 2016
Copyright c By Daniel W. Gaddes 2016

_______

Not being in the line of fire lately (bad joke for Li batteries), I was unaware of any pulse charging studies. I did not note any reference to any commercial implementation. Do you? With the pervasiveness of Li batteries in products from phones to cars, all including chargers, surprising none of those marketing departments has jumped into pulse charging.

So again, which source or reference put you on to it and what benefits do you expect and how exactly do you go about choosing and implementing the algorithm?
bi

[Turion]

Google it. It’s called research. Here is one article that came up immediately.

https://iopscience.iop.org/article/1...6/817/1/012008

[bistander]

... Pulse charging methods has been developed as one of the fast charging methods for Lithium ion battery. This technique applies the continuous constant current and pulse with certain pulse width until the battery fully charged.
...

Hi Turion,
I have experience with Lithium ion batteries, both hands on and academic (theoretical, educational, research, etc.) and never encountered a fast charging method using pulses. Please provide sources or references for your claim.
Thanks.
bi

[Turion]

It will be a while

[bistander]

All the coils had 3 strands of 1000 ft in parallel. AWG #23 Rotor at 3,000 rpm, or as close as I could get. But thanks for the questions. You made me realize I screwed up. When I went from 4 magnets to two, I took out every other one, which means both the magnets left in the rotor were the same polarity. I need to flip one of them and retest. That will take a bit. Need to get the magnet out of the rotor, and I'm pretty sure my partner took all 12 coils that were for the machine. There are some test coils, but I'm not sure what the wiring configuration is. Hope one is just 3 strands. I know there are 6 strand, 12 strand and 24 strand, and it changes the capacitance and output when you start connecting parallel wires in series.

Reconnecting the 6 strand into a 3 strand equivalent should give close enough data. I imagine difference in capacitance compared to coil wound as 3 strand would be of minimal importance, especially on no-load. Please measure and record coil resistance.

In my last paragraph I meant that under the exact same conditions (except core material) the old and new cores are not the same. The new permalloy cores only put out 75% of what the old iron cores did. That isn't what I was led to believe.

If the coils (windings) were exactly the same ( T/c, wire gauge & R), then different cores were responsible for the 75%. Not sure what old iron you used. Getting saturation flux density data on it might be lost cause. You like to see it on your bench, so test old iron core vs permalloy core for flux from the magnet. If permalloy core showed only 75% Voc, then logically it was putting only 75% of the flux thru the core. I think the amount of flux from the magnet thru the core is proportional to the pull force, or vise versa. But measuring the pull force to separate the magnet from the core, straight on, separated by a plastic shim thick as your air gap, will show you a big clue. Pull test shouldn't be affected by coils, if oc.

Or perhaps, instead, or in addition, use the plastic shim, stick a core to the magnet and read the flux at opposite end of the core with gaussmeter.

I didn't record AC watts input. I wasn't running my generator. I was using my little test rotors on my bench grinder.
https://youtube.com/shorts/zU2Mh3Q184M?feature=share

I thought that bit of data would be interesting. Please use your killawatt meter on the wall plug for the Variac and take a second each test to note the watts from the grid. All else equal, what's the difference no-load supply power between 2 magnets, 4 magnets, iron core, permalloy core, no core?
bi

[Turion]

All the coils had 3 strands of 1000 ft in parallel. AWG #23 Rotor at 3,000 rpm, or as close as I could get. But thanks for the questions. You made me realize I screwed up. When I went from 4 magnets to two, I took out every other one, which means both the magnets left in the rotor were the same polarity. I need to flip one of them and retest. That will take a bit. Need to get the magnet out of the rotor, and I'm pretty sure my partner took all 12 coils that were for the machine. There are some test coils, but I'm not sure what the wiring configuration is. Hope one is just 3 strands. I know there are 6 strand, 12 strand and 24 strand, and it changes the capacitance and output when you start connecting parallel wires in series.

In my last paragraph I meant that under the exact same conditions (except core material) the old and new cores are not the same. The new permalloy cores only put out 75% of what the old iron cores did. That isn't what I was led to believe.

I didn't record AC watts input. I wasn't running my generator. I was using my little test rotors on my bench grinder.
https://youtube.com/shorts/zU2Mh3Q184M?feature=share

[bistander]

I spent the last week figuring out what the issues were. Shorted coil, bad connections, etc. But here is some more data from testing on my little 4 magnet rotor setup. I was STILL CONCERNED about output of coils on the big machine, even when I figured out and fixed all the problems, but it is what it is. Plus right now it has to be disassembled to get the magnet that slipped out of the rotor unstuck from the coil core. Let someone else wast two days doing that. I'm busy.

Its output will not be improved without the addition of more coils or a different core material. Using my bench grinder running on a Variac so I could adjust the motor and control the RPM, I tested the OLD coils and the NEW coils with two magnets in the rotor and four magnets on the rotor to try and get some idea of what the output should be with 10 magnets on the rotor, or 20, or 24. Here is what I saw.

Old coil with 4 magnets 58.6 volts. New coil with 4 magnets 44.5 volts. 76% of the old coil output for the new coil
Old coil with 2 magnets 47.7 volts. New coil with 2 magnets 34.4 volts. 72% of the old coil output for the new coil

Adding 2 magnets to the "OLD" setup gave me an 8 % increase
Adding 2 magnets to the "New" setup gave me a 7.7 % increase
So roughly the same.

I don't like to extrapolate numbers because that's guess work and not bench research. BUT I also have actual RESULTS from testing that indicate some of this might be factual
If the old coil was outputting 120-130 volts with a 12 magnet rotor (it was) I should only expect the NEW coils to output around 88-96 volts volts (at 74% of the old coils as an average) with a 12 magnet rotor. At 74% of what the old machine put out, that's somewhere between 1,350 to 1500 watts on an input of between 300-400 watts. It won't output as much as the old machine, but it won't overheat either. Right now, for a prototype, that is more important to me than output. Let someone else spend years searching for the perfect core. I'm done with this project.

Interestingly, the 20 magnet rotor was showing a coil output of 96 volts (new coils) from a coil PAIR. which is two coils in parallel. That 96 volts would be roughly 75%, so not out of line with my testing (except I had 20 magnets on the rotor, not 12)
The 10 magnet rotor, before it locked up, was only showing an output of 36 volts. Still don't know why.

My guys, who did all the testing of the different core materials on the old clunker machine, told me the new coils were almost the exact same as the old coils. Not really. Not by a long ways.

Thanks. Nice to see some data.

Please define more precisely old and new coils in terms of number of turns per coil (or coil dimensions and wire length), wire gauge, core dimensions and material.

What was the RPM for each of the 2 & 4 magnet tests? We're the magnets always the same and N, S or N, S, N, S?

Your last paragraph indicates you think coils are different. Do you mean different in size, # of turns, wire gauge or different in core material and/or core size.

Just for kicks, did you happen to record the AC watts input for the 2 & 4 magnet tests?
bi

[Turion]

I spent the last week figuring out what the issues were. Shorted coil, bad connections, etc. But here is some more data from testing on my little 4 magnet rotor setup. I was STILL CONCERNED about output of coils on the big machine, even when I figured out and fixed all the problems, but it is what it is. Plus right now it has to be disassembled to get the magnet that slipped out of the rotor unstuck from the coil core. Let someone else wast two days doing that. I'm busy.

Its output will not be improved without the addition of more coils or a different core material. Using my bench grinder running on a Variac so I could adjust the motor and control the RPM, I tested the OLD coils and the NEW coils with two magnets in the rotor and four magnets on the rotor to try and get some idea of what the output should be with 10 magnets on the rotor, or 20, or 24. Here is what I saw.

Old coil with 4 magnets 58.6 volts. New coil with 4 magnets 44.5 volts. 76% of the old coil output for the new coil
Old coil with 2 magnets 47.7 volts. New coil with 2 magnets 34.4 volts. 72% of the old coil output for the new coil

Adding 2 magnets to the "OLD" setup gave me an 8 % increase
Adding 2 magnets to the "New" setup gave me a 7.7 % increase
So roughly the same.

I don't like to extrapolate numbers because that's guess work and not bench research. BUT I also have actual RESULTS from testing that indicate some of this might be factual
If the old coil was outputting 120-130 volts with a 12 magnet rotor (it was) I should only expect the NEW coils to output around 88-96 volts volts (at 74% of the old coils as an average) with a 12 magnet rotor. At 74% of what the old machine put out, that's somewhere between 1,350 to 1500 watts on an input of between 300-400 watts. It won't output as much as the old machine, but it won't overheat either. Right now, for a prototype, that is more important to me than output. Let someone else spend years searching for the perfect core. I'm done with this project.

Interestingly, the 20 magnet rotor was showing a coil output of 96 volts (new coils) from a coil PAIR. which is two coils in parallel. That 96 volts would be roughly 75%, so not out of line with my testing (except I had 20 magnets on the rotor, not 12)
The 10 magnet rotor, before it locked up, was only showing an output of 36 volts. Still don't know why.

My guys, who did all the testing of the different core materials on the old clunker machine, told me the new coils were almost the exact same as the old coils. Not really. Not by a long ways.

[bistander]

....
By the way, my partners picked the NEW generator and the OLD generator up today. Taking it to the city where they will fine tune it and FINALLY have it tested. They want to get the old one up and running also. Having a new shaft made and new bearings. ...

Hi Turion,
Interesting!
"fine tune it and FINALLY have it tested" Basic functionality has yet to be demonstrated. You really think you can hand it off and someone else can do this?
Good luck.
bi