Turnigy Multistar - lightweight 10C cells
- Thread starter miuan
- Start date
Addy
100 W
Ianhill said:
I have two diplomas as an EE Technologist in Computer Control and Industrial Power. I work as an electrical designer.
All I'm saying is that with adequate insulation and system design, you could build very high voltage battery packs. You haven't disproven this and your information about internal cell insulation is irrelevant.
Addy
100 W
Ianhill said:
Yes, the voltages at the cell terminals with respect to ground/pack negative increase as you add more cells in series. The voltage across a single cell does not change as you add other cells in series with it.
Here is a paper about high voltage battery packs for EVs. They talk about packs with 100-200 cells.
https://cds.linear.com/docs/en/article/EDN M Kultgen LTC6802.pdf
Please point out the part where they talk about dielectric breakdown. Just kidding, it's not in there, because it's irrelevant.
Addy
100 W
Ianhill said:
I quieted down because I thought you agreed finally. I'm not saying dielectric breakdown isn't real, perhaps you should sharpen up your English skills. I said it is irrelevant, because it isn't going to happen inside the cell, because the cell voltage doesn't change when adding more cells in series. If you were to take a basic electronics course you would learn that voltage does not go "through" anything. Current flows through, not voltage.
I will agree that people who don't understand the complications and risks of high voltage battery packs should avoid them for safety reasons.
Ohbse
10 kW
Ianhill said:
Sorry Ian, think you're the one slinging nonsense here. What you're saying flies in the face of every production high voltage pack. As previously mentioned Chris Jones runs a 700v pack from hobby cells with no more inter cell insulation than your average 4s hobby pack. Of course he pays a great deal of attention to insulation between areas with higher potential differences than a single cell.
Easiest way to disprove what you're saying, there's lots of videos of people daisy chaining simple 9v alkaline batteries together up to multi KV levels. These of course produce great spark shows, but only from pole to pole. No cell level breakdown to be seen.
That's not to say that high voltage packs are a good idea in a salty wet environment. The example of a tesla charging connector getting wet is not really relevant to something being immersed in salt water. There's no compelling reason to go over 20s for less than 50kw of output IMO. That power limit for 72v will inevitably be raised as more mature power switching comes along.
Ohbse
10 kW
[youtube]8hwLHdBTQ7s[/youtube]
Here's a guy running 2kv for extended periods of time with 244 pp3's in series
While I don't claim to be an expert, I think you've got a fundamental misunderstanding of how voltage works as what you're describing is contrary to not just my own observations but observable reality with pack designs by people much smarter than me. Tesla run 400v using conventional 18650's. Chris has built multiple 700v packs using hobby cells that are actually touching each other with only a thin film of separator. He's run these extremely hard with no voltage related breakdowns.
I don't disagree that the terms and processes you're describing exist and are relevant to pack design, but they're not a factor in single cell series connections from everything I can find. If you can find a reference I would be very happy to be proven incorrect, but at the moment the evidence says you're mistaken.
Here's a guy running 2kv for extended periods of time with 244 pp3's in series
While I don't claim to be an expert, I think you've got a fundamental misunderstanding of how voltage works as what you're describing is contrary to not just my own observations but observable reality with pack designs by people much smarter than me. Tesla run 400v using conventional 18650's. Chris has built multiple 700v packs using hobby cells that are actually touching each other with only a thin film of separator. He's run these extremely hard with no voltage related breakdowns.
I don't disagree that the terms and processes you're describing exist and are relevant to pack design, but they're not a factor in single cell series connections from everything I can find. If you can find a reference I would be very happy to be proven incorrect, but at the moment the evidence says you're mistaken.
Ianhill
1 MW
- Joined
- Sep 25, 2015
- Messages
- 2,871
A typical pp3 is 9.5v so 244 times that gives 2318vdc and discharge rate at 30ma so no wonder they are poping and hissing the electrolyte is boiling inside the cell, having discharged at well over a few amp for that arc,
At the end he show ms the temp in the cells and the fact one has failed to a dead short is telling you that the limit is reached.
Each brand to their own aswell as cell design so the voltage a pp3 fails at is not necessary where a lipo will fail at they are two different designs, I can not say how many more he could of add to his string if any or where his missing 300vdc was that's on average each cell 1.5v lower than fully charged so barley 20% capacity left, it's clear to me that leakage was occuring at that high level due to physical limitations,
The most I can find is 700vdc pack with 20ah lithuim iron phosphate cells in a 170s pack, so is this the safe practical limit ?
At the end he show ms the temp in the cells and the fact one has failed to a dead short is telling you that the limit is reached.
Each brand to their own aswell as cell design so the voltage a pp3 fails at is not necessary where a lipo will fail at they are two different designs, I can not say how many more he could of add to his string if any or where his missing 300vdc was that's on average each cell 1.5v lower than fully charged so barley 20% capacity left, it's clear to me that leakage was occuring at that high level due to physical limitations,
The most I can find is 700vdc pack with 20ah lithuim iron phosphate cells in a 170s pack, so is this the safe practical limit ?
ElectricGod
10 MW
I started on page 19...so I haven't read everything.
Any single cell as long as it is only connected in series with the two cells adjacent to it whether that's in a string of 5 cells or 500 cells, will only ever see its individual cells voltage. How can it be different than this? That's assuming a few things. It's a simple voltage divider. 5 equally valued resistors across a voltage will show an equal voltage drop across each resistor. If that voltage is 5 volts or 5000 volts...doesn't matter. Make that 5000 resistors and the same thing happens, just each resistor shows a smaller voltage drop across each one. Individual cells in a 500S pack will work identically as 500 resistors in series. Each cell will be equivalent to a resistor in series with lots of other resistors and the total voltage will be divided equally among all of them...providing the internal resistance of each cell is the same. And that's where the rub lies...cells with differing Ir can potentially have more or less voltage across them. It's still just a giant voltage divider. Fortunately there is this lovely invention called a BMS that will keep all the cells in balance so that this can't happen despite varying Ir. No individual cell is going to see the voltage of more than its own voltage just as a simple voltage divider. And the BMS is going to prevent that from happening while charging. Most decent BMS do on the fly balancing so that if a cell shows too much voltage, it is discharged down to a safe level. Of course there are limits to everything and these safety functions can be overwhelmed. A BMS for these voltages had better have a way to shut off it's mosfets or IGBT's if cell imbalances get too out of whack. This sort of a pack is not unusual...Tesla and many EVr's built very high voltage packs and they do work reliably...if the proper precautions are made.
Maybe those 1200 volt packs had mitigating reasons for the cell failure that wasn't specific to overcoming cell dielectric? I wonder what else was happening with that pack? Ian, find the video and post it here please.
It looks like Tesla and Leaf are both running at 400 volts or so. There may be something to this or else why wouldn't Tesla make the P100S run at 500 volts and then make an insanely ludicrously fast car? Or maybe they are working on that. LOL!
My greatest concern with salt water and voltage of any kind is the electrocution factor. Salt water is a really good conductor...don't ever let that much voltage come in contact with the water You will kill everyone near by. Even 48 volts at the wattage you want in salt water is going to be a deadly combination. Furthermore salt water is corrosive. It's going to eventually invade any kind of metal box you use. It's just a matter of time. Any kind of materials that can soak up water need to never EVER get wet at the voltages you want. That's an instant path of low resistance for current to pass through and now you will have the opportunity for cells to see much, MUCH more than their own voltage. I'd pot the tops of all the cells in epoxy or silicon after the pack was built. I'd put a heat conductive layer between each cell like the Volt packs have. This not only allows a heat path in the middle of the pack, but it also creates an isolation layer between cells in case one does go thermal. Then I'd put the whole thing inside a vacuum sealed bag and that inside 2 more vacuum sealed bags and then that inside a fiber glass box that was water tight up to 50 feet depth. Then all wiring from the pack to the controller and to the motor and contactor or anywhere that the high voltage went would get potted and then vacuum sealed multiple times. Assume when you are done that you will never service anything ever again because you won't be able to without destroying lots of stuff to get at whatever failed. You can't afford any water penetration into anything or else you die.
Any single cell as long as it is only connected in series with the two cells adjacent to it whether that's in a string of 5 cells or 500 cells, will only ever see its individual cells voltage. How can it be different than this? That's assuming a few things. It's a simple voltage divider. 5 equally valued resistors across a voltage will show an equal voltage drop across each resistor. If that voltage is 5 volts or 5000 volts...doesn't matter. Make that 5000 resistors and the same thing happens, just each resistor shows a smaller voltage drop across each one. Individual cells in a 500S pack will work identically as 500 resistors in series. Each cell will be equivalent to a resistor in series with lots of other resistors and the total voltage will be divided equally among all of them...providing the internal resistance of each cell is the same. And that's where the rub lies...cells with differing Ir can potentially have more or less voltage across them. It's still just a giant voltage divider. Fortunately there is this lovely invention called a BMS that will keep all the cells in balance so that this can't happen despite varying Ir. No individual cell is going to see the voltage of more than its own voltage just as a simple voltage divider. And the BMS is going to prevent that from happening while charging. Most decent BMS do on the fly balancing so that if a cell shows too much voltage, it is discharged down to a safe level. Of course there are limits to everything and these safety functions can be overwhelmed. A BMS for these voltages had better have a way to shut off it's mosfets or IGBT's if cell imbalances get too out of whack. This sort of a pack is not unusual...Tesla and many EVr's built very high voltage packs and they do work reliably...if the proper precautions are made.
Maybe those 1200 volt packs had mitigating reasons for the cell failure that wasn't specific to overcoming cell dielectric? I wonder what else was happening with that pack? Ian, find the video and post it here please.
It looks like Tesla and Leaf are both running at 400 volts or so. There may be something to this or else why wouldn't Tesla make the P100S run at 500 volts and then make an insanely ludicrously fast car? Or maybe they are working on that. LOL!
My greatest concern with salt water and voltage of any kind is the electrocution factor. Salt water is a really good conductor...don't ever let that much voltage come in contact with the water You will kill everyone near by. Even 48 volts at the wattage you want in salt water is going to be a deadly combination. Furthermore salt water is corrosive. It's going to eventually invade any kind of metal box you use. It's just a matter of time. Any kind of materials that can soak up water need to never EVER get wet at the voltages you want. That's an instant path of low resistance for current to pass through and now you will have the opportunity for cells to see much, MUCH more than their own voltage. I'd pot the tops of all the cells in epoxy or silicon after the pack was built. I'd put a heat conductive layer between each cell like the Volt packs have. This not only allows a heat path in the middle of the pack, but it also creates an isolation layer between cells in case one does go thermal. Then I'd put the whole thing inside a vacuum sealed bag and that inside 2 more vacuum sealed bags and then that inside a fiber glass box that was water tight up to 50 feet depth. Then all wiring from the pack to the controller and to the motor and contactor or anywhere that the high voltage went would get potted and then vacuum sealed multiple times. Assume when you are done that you will never service anything ever again because you won't be able to without destroying lots of stuff to get at whatever failed. You can't afford any water penetration into anything or else you die.
ElectricGod
10 MW
Ianhill said:
I bet he was leaking voltage through the metal shells of the 9v batteries. If he put each battery in it's own heat shrink tube so that the metal shells couldn't touch, that voltage may have been a good bit higher. Also it was lots of used cell that are partly discharged. I bet none of them were at 9 volts. The popping and hissing may not have had anything to do with electrolyte boiling. At those voltages, current will pass quite nicely across the air. That's why I would want to electrically isolate all the batteries from each other with individual heat shrink tubes. AT 400 volts Like the Tesla uses, if the cells were not isolated, they would arc to each other easily. Even a single 9 volt battery can sustain a tiny arc. At 1200 volts...well that's just a much bigger arc. Probably the dead battery at the end of the video was do to the fact that he was dead shorting the whole string of batteries. At those voltages and under dead short conditions...things ARE going to fail.
If this 9 volt battery pack experiment video is what you were referring to earlier when you mentioned a 1200 volt pack, then yes, I can totally see why the cells ran down. Nothing is electrically isolated from anything else in that thing.
Addy
100 W
Ianhill said:
Voltage doesn't "enter" the cell. When you're measuring the pack voltage you're measuring the sum of all the cells added together. It doesn't change the voltage of each cell, and each individual cell doesn't see anything besides it's own terminal voltage.
ElectricGod
10 MW
It's just additive voltage...
cell1+cell2+cell3 = total pack voltage
That does not mean that cell1 in circuit "sees" the voltage of cell 2 and 3. It only sees it's own voltage...assuming that there is no other current path. That's why I said there had to be other mitigating circumstances that caused cells to run down.
It's those other current paths like the shells on those 9v batteries popping tiny arcs back and forth that are a problem. That guy in the video was lucky that none of his batteries had bad internal insulation or that two that did were no where near each other in the whole bank or else he would have gotten a nice big spark like he did with the alligator clip.
You do realize that no one NOT EVER designed a 9v battery to run in a bank of 244 of them in series. This is completely outside the design limits for any safety margins built into a typical 9 volt battery. So very likely the batteries shells were a convenient new current path at those voltages that allowed the batteries to run down. If they were all electrically isolated from each other, I seriously doubt they would run down at all or at least not any faster than they would individually sitting on a shelf.
cell1+cell2+cell3 = total pack voltage
That does not mean that cell1 in circuit "sees" the voltage of cell 2 and 3. It only sees it's own voltage...assuming that there is no other current path. That's why I said there had to be other mitigating circumstances that caused cells to run down.
It's those other current paths like the shells on those 9v batteries popping tiny arcs back and forth that are a problem. That guy in the video was lucky that none of his batteries had bad internal insulation or that two that did were no where near each other in the whole bank or else he would have gotten a nice big spark like he did with the alligator clip.
You do realize that no one NOT EVER designed a 9v battery to run in a bank of 244 of them in series. This is completely outside the design limits for any safety margins built into a typical 9 volt battery. So very likely the batteries shells were a convenient new current path at those voltages that allowed the batteries to run down. If they were all electrically isolated from each other, I seriously doubt they would run down at all or at least not any faster than they would individually sitting on a shelf.
Ianhill
1 MW
- Joined
- Sep 25, 2015
- Messages
- 2,871
ElectricGod said:
Between the two half's there would of been around 1000vdc potential difference the fact it made a clicking sound when it touched showed there was a small dielectric leakage path occuring throught the shells but with poor conductivity, if the voltage had gotten higher then the two half's would of sparked big time.
I can see that when we series a battery we raise the next cells negative electrode to a higher potential there's always 4.2v across the cell and that's what stops the dielectric brakedown between electrodes of each battery so I do have to eat my words on that one and I have learnt something once again shut my mouth.
I'll say it again I was wrong on dielectric leakage between electrodes I'll delete the comments not so I don't look a wanker but not to give the wrong idea to someone, damn I'm eating my words on this one but your a silky twat for riding 800vdc in the sea I got to say that one.
I worked it out for myself I'm chatting shit on dielectric it has no effect due to the 4.2v potential difference between electrodes, when we series battery all we do is raise the potential on the negative electrode so the only leakage paths to worry about is between terminals/ cell housings of much differing voltage and the case that contains them having dielectric breakdown and allowing leakage from the high voltage side to low.
rainmaking
1 mW
- Joined
- Aug 20, 2017
- Messages
- 13
I'm sure sorry I brought up the Tesla charge connector....now it appears I don't even know what the difference between salt and rain water is even when I state it's a completely different environment. BTW Tesla is likely using 400V because it is inconvenient to boost high current charge voltages higher than what you get rectified from 240VAC. The prius uses up to 700VDC in its drivetrain. Tesla batteries are split into 24V modules that are series'd together. the car could quite possibly get submerged in saltwater and the occupants will be isolated from the major current flow even if they are touching the saltwater.
In my case I think it would be counter productive to pot each cell top. I think in the catastrophic event of sinking with a simultaneous battery enclosure failure that I would want the cells to have every path possible to discharge. Keep in mind the entire outside of the jet ski is fiberglass, i.e. non conductive. the drive shaft and some control wires are the only conductive things that penetrate the enclosure. I will consider how to protect these items from transmitting high voltages outside the ski hull but I am not going to consider potting the entire pack. There is an electric jet ski on the market as are quite a number of maritime energy solutions in this voltage range.
I appreciate the concerns and feedback. I posted on this thread specifically to inquire about these multistar bricks but got more than I bargained for. I don't really want to have a debate about sealed pack design. If anyone has specific experience I will like to read it.
Ian the bike video is at the end of jonesc's build thread, the link is about 15 posts up. He hit about 160kW a few times which is in the normal range for a superbike. I suspect his IR losses were pretty high at those peaks.
In my case I think it would be counter productive to pot each cell top. I think in the catastrophic event of sinking with a simultaneous battery enclosure failure that I would want the cells to have every path possible to discharge. Keep in mind the entire outside of the jet ski is fiberglass, i.e. non conductive. the drive shaft and some control wires are the only conductive things that penetrate the enclosure. I will consider how to protect these items from transmitting high voltages outside the ski hull but I am not going to consider potting the entire pack. There is an electric jet ski on the market as are quite a number of maritime energy solutions in this voltage range.
I appreciate the concerns and feedback. I posted on this thread specifically to inquire about these multistar bricks but got more than I bargained for. I don't really want to have a debate about sealed pack design. If anyone has specific experience I will like to read it.
Ian the bike video is at the end of jonesc's build thread, the link is about 15 posts up. He hit about 160kW a few times which is in the normal range for a superbike. I suspect his IR losses were pretty high at those peaks.
ElectricGod
10 MW
rainmaking said:
Is 240 volts AC rectified 400 volts DC? I don't honestly know...Id' have to go google that, but if it is then yeah that would 100% explain the 400 volt typical voltage for electric cars. Hybrids don't count. You don't charge the battery off your AC mains. You charge them off a purpose built generator incorporated into the car. IE: Prius which is a total POS in my opinion can run at any battery voltage Toyota chooses. Anytime you boost voltage whether that's with a giant transformer or via a step up switching power supply, you are always trading current for more voltage. When you step down voltage, you also gain current. Since your AC mains are typically a couple hundred amps, it is logical that Tesla would try to take advantage of that fact with as minimal losses as possible. They would want to rectify the AC mains directly to DC and then not mess with it much other than to manage current loads so that they don't overheat your mains wires or blow your main breaker at your meter.
Tesla splits up their packs into chunks for really good reasons. Each chunk can be individually replaced if it goes bad. Those chunks are the basic building block for any capacity or voltage they could ever want to build. Tesla has some pretty smart folks working there. They tend to think of this sort of thing.
Potting...yeah not the critical element, just keep inside that pack uber dry and make sure there is no chance of a current path into the water. How do you suppose people get electrocuted by their blow drier when they are in the bathtub? It's the voltage potential in the blow drier traveling to a lower voltage potential that happens to also pass through the person. In the case of your jet ski, you have 800 volts at a zillion amps potential spreading outwards in every direction that happens to include you and anyone else nearby in the water. Zap...your are dead...just don't let that happen.
You would want a way to vent gasses out of the battery box, but you absolutely do not want water penetration...water = corrosion = bad. I would not trust the jest ski hull to be my only insulator from all that voltage. I bet the commercially available units don't either! I want to live long enough to ride my electric beast more than once! There's nothing wrong with an electric jet ski...it just has implications to work out.
rainmaking
1 mW
- Joined
- Aug 20, 2017
- Messages
- 13
ElectricGod said:
Mostly agree. Like a story passed down from multiple tellers, somehow my 96S has turned into 800V. I wish I knew what chemistry caused that
.I'm also not sure being in saltwater with a bunch of cells discharging would be deadly. Why would the current not travel in the lowest resistance path, closest to the highest potential conductors? Not that it's not worth precaution, and of course the goal is to NEVER let any high voltage conductors get wet. I think the higher risk is working on the system w/o proper precaution.
Venting needs investigation. I would want a gore vent on it to minimize humidity. Also need to vent a blown cell without enclosure failure. Like a pressure relief valve but a pretty low pressure.
rainmaking
1 mW
- Joined
- Aug 20, 2017
- Messages
- 13
Oh, and I would think grounded plumbing has something to do with hair dryer electrocutions.
Sunder
10 MW
ElectricGod said:
Huh. 240vac RMS rectified is 336vdc. 220v stepped down is 308v
I was wondering why my Outlander had a 300v pack. Would be very efficient to step down (if it's stepped down at all).
Hillhater
100 TW
If you blow a cell in that situation, you will need an ejector seat !
when lipo blows,..its inevitably a firey event that snowballs fiercely..practically an explosion !
