Making a bulk charger.
- Thread starter Sigma
- Start date
heathyoung
100 kW
What are your abilities? You could use a number of meanwell power supplies in series, or make your own switchmode buck power supply (I'm in the middle of doing this at the moment for a vectrix charger 150V@10A)
Philistine
100 kW
- Joined
- Aug 15, 2010
- Messages
- 1,736
I realise $250 is nothing to be sneezed at, and not everyone has it lying around, but personally I would just buy a charger from BMS battery. http://www.bmsbattery.com/alloy-she...ifepo4li-ionlead-acid-battery-ev-charger.html
They can make it exactly to your chosen voltage/current. Obviously this will be more expensive than building your own (potentially as much as $200 more), but in my opinion when I list the things I am willing to save money on by going ghetto and homebrew, charging my Lipo is not one of them (this also solves your concerns re being in one unit). I totally understand why people want to save cost everywhere they can, but I put the quality and reliability of my charger up there in importance with my brakes and tyres in being worth spending money on. Just my two cents. That's why I don't understand why people muck around with Meanwells, they are not even much cheaper than a BMS charger. At least with modded server PSU's I can appreciate the massive price difference and current potential.
I have the 1200W BMS charger and it has run a treat for about 8 months of daily use.
They can make it exactly to your chosen voltage/current. Obviously this will be more expensive than building your own (potentially as much as $200 more), but in my opinion when I list the things I am willing to save money on by going ghetto and homebrew, charging my Lipo is not one of them (this also solves your concerns re being in one unit). I totally understand why people want to save cost everywhere they can, but I put the quality and reliability of my charger up there in importance with my brakes and tyres in being worth spending money on. Just my two cents. That's why I don't understand why people muck around with Meanwells, they are not even much cheaper than a BMS charger. At least with modded server PSU's I can appreciate the massive price difference and current potential.
I have the 1200W BMS charger and it has run a treat for about 8 months of daily use.
Philistine
100 kW
- Joined
- Aug 15, 2010
- Messages
- 1,736
Quite a few people have bought them on here, Doc Bass tricked one out with variable voltage pots etc. (look for his thread, he doesn't refer to it as a BMS charger though I think). As I said, my 1200W has run like magic for about 8 months.
It is big (she box sized), but has a digital display of volts and amps, and volts and amps can be adjusted with a pot. The postage is not cheap but not crazy, and delivery time is something like 4 to 6 weeks (at least to Aus it was).
Do a search on here for BMS and King or Kingpan Chargers. Most people are happy with them, I think there has only been one or two failures. You wouldn't want to carry the thing around, it is big and bulky, and I suspect would be a bit fragile, but for a home charger it is (in my opinion) the obvious bulk charging choice. The only reason I can see for not going that route is because you literally can't scratch the $200 price difference between that and a homebrew PSU job, but personally I am happy to find that scratch for peace of mind. Having said that, letting me at any meaningful electrical work is like letting John Wayne Gacey be the Clown at your son's birthday party.....
It is big (she box sized), but has a digital display of volts and amps, and volts and amps can be adjusted with a pot. The postage is not cheap but not crazy, and delivery time is something like 4 to 6 weeks (at least to Aus it was).
Do a search on here for BMS and King or Kingpan Chargers. Most people are happy with them, I think there has only been one or two failures. You wouldn't want to carry the thing around, it is big and bulky, and I suspect would be a bit fragile, but for a home charger it is (in my opinion) the obvious bulk charging choice. The only reason I can see for not going that route is because you literally can't scratch the $200 price difference between that and a homebrew PSU job, but personally I am happy to find that scratch for peace of mind. Having said that, letting me at any meaningful electrical work is like letting John Wayne Gacey be the Clown at your son's birthday party.....
Ricky_nz
10 kW
Definitely safer then rolling your own mains powered switch mode power supply!
100+Vdc is starting to get rather hazardous too.
100+Vdc is starting to get rather hazardous too.
Alright so I want to confirm these settings to make sure I put the right values before ordering. So I have a 30S2P each pack is 6cells at 22.2 Volts. With a full charge at 25.2 Volts. (ZIPPY Flightmax 5000mAh 6S1P 25C http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=8586)
Battery type: LIPO
Charge voltage: 126V
Charge current: 10A
AC voltage 110-120 Volts
AC plug: standard US/Cdn plug
DC plug: JST-XH?
Battery type: LIPO
Charge voltage: 126V
Charge current: 10A
AC voltage 110-120 Volts
AC plug: standard US/Cdn plug
DC plug: JST-XH?
I'm working on such a charger atm. The only problem is it will be rather unefficient at 110vAC, mine is designed and works at 220-240vAC.
It's a simple, yet very efficient charger. It uses a AVR to sense ZVS (AC directly into mcu is cool
), and use either of the two mosfets to chop the AC. A high powered ultrafast diode will maybe be replaced by a another mosfet - but not yet.
It uses a current sensor IC (able to handle 200A) to controll the duty cycle. It adjusts duty cycle based on current feedback. This current sensor can be "re used" as a AH meter while riding.
As of now I've reached a limit on 1300W, magnetics does in theory - and practice stop being very efficient here. 5 minute bouts at 1700W seems to go ok tho.
It's not heavy, all in all it weighs in at just shy of a kg. More power can be squeezed out adding snubbers to fets and driving them to higher frequency.
Power factor, you ask? Well, it's complicated - that's my reply xD. I haven't got the equipment on how bad it is atm - but 85% (given 75V pack voltage on 220VAC grid) of the time it is drawing constant power from the grid. The current drawn looks like a square wave with a inverted sinus on top. A more exotic algorithm could pull sine wave current too. A buck boost converter would enable me to get 0.99 power factor, but at higher components cost.
If this sounds like greek to you and you don't know where to start, I suggest buying something ready made.
It's a simple, yet very efficient charger. It uses a AVR to sense ZVS (AC directly into mcu is cool
), and use either of the two mosfets to chop the AC. A high powered ultrafast diode will maybe be replaced by a another mosfet - but not yet.It uses a current sensor IC (able to handle 200A) to controll the duty cycle. It adjusts duty cycle based on current feedback. This current sensor can be "re used" as a AH meter while riding.
As of now I've reached a limit on 1300W, magnetics does in theory - and practice stop being very efficient here. 5 minute bouts at 1700W seems to go ok tho.
It's not heavy, all in all it weighs in at just shy of a kg. More power can be squeezed out adding snubbers to fets and driving them to higher frequency.
Power factor, you ask? Well, it's complicated - that's my reply xD. I haven't got the equipment on how bad it is atm - but 85% (given 75V pack voltage on 220VAC grid) of the time it is drawing constant power from the grid. The current drawn looks like a square wave with a inverted sinus on top. A more exotic algorithm could pull sine wave current too. A buck boost converter would enable me to get 0.99 power factor, but at higher components cost.
If this sounds like greek to you and you don't know where to start, I suggest buying something ready made.
voicecoils
1 MW
Sigma said:
Having a bulk charger charge to 4.2v per cell is generally bad practice, unless you have full BMS onboard. I would have the bulk charger set to to 124v (4.13v/cell) or 125v (4.17v/cell) to give a margin of safety if cells go out of balance.
Also, JST-XH is not an appropriate plug for your DC battery connection, especially at 10A charge current. That's what the Zippy balance taps use and you need to charge through the main power leads. To connect the pack to your charger, you'll need either the 4mm bullet connectors which you can buy at hobbyking, or change to a different DC power connector such as an Anderson PP45 or SB50.
Good luck!
Teh Stork said:I'm working on such a charger atm. The only problem is it will be rather unefficient at 110vAC, mine is designed and works at 220-240vAC.
It's a simple, yet very efficient charger. It uses a AVR to sense ZVS (AC directly into mcu is cool
It uses a current sensor IC (able to handle 200A) to controll the duty cycle. It adjusts duty cycle based on current feedback. This current sensor can be "re used" as a AH meter while riding.
As of now I've reached a limit on 1300W, magnetics does in theory - and practice stop being very efficient here. 5 minute bouts at 1700W seems to go ok tho.
It's not heavy, all in all it weighs in at just shy of a kg. More power can be squeezed out adding snubbers to fets and driving them to higher frequency.
Power factor, you ask? Well, it's complicated - that's my reply xD. I haven't got the equipment on how bad it is atm - but 85% (given 75V pack voltage on 220VAC grid) of the time it is drawing constant power from the grid. The current drawn looks like a square wave with a inverted sinus on top. A more exotic algorithm could pull sine wave current too. A buck boost converter would enable me to get 0.99 power factor, but at higher components cost.
If this sounds like greek to you and you don't know where to start, I suggest buying something ready made.
Your charger sounds very interesting so how much is it going to cost you to build it all, do you have any pics?
voicecoils said:
Ah typo, I ment to put the 4mm bullet connectors. Your right, charging to 4.24 volts is better, what voltage value would be the most efficient to keep a good cycle life with the batteries?
Sigma said:
Build cost is about 20-30 dollars in fets (could be done for 10 dollars - but with significant heat production and ~10% efficiency loss), 5 dollar mcu,, 4 dollar current IC sensor, 5 dollar pcb, Huge ass toroid - had it lying around ~5-10 dollar, 30meter enammeled AWG18 copper wire (using 5 parallelled strands to get resistance down), 2x mosfet driver 4 dollars, dollar fuse and about a dollar worth of resistors, tactile buttons and capacitors. Working hours, so far; ~10 ish. The project is still on the evaluation-board stage, but has charged my pack successfully some times now.
I plan to implement 3-5 thermistors and a "is the pack in balance" check in the form of a middle point test point. I'm also adding sleep functionality, enabling it to dwell at 1uA draw. Current draw from the battery pack is meant to be done through the current sensor, so the charger will have a idea of how much "fuel" is used.
As for the price once I'm done, not sure - but I'm guessing around 100 dollars would be a ballpark figure. I won't ruin my reputation by sending out a product killing batterypacks - so I plan on 'doing this right' the first time.
It slowly evolves to something like a child of a threesome starring a BMS, Cycle Analyst and a charger xD
Teh Stork said:
Alright good luck to you, would you happen to have an ETA of when yours will be done? lol I dont want to go spend $250 + about $100 shipping on that charger and then find out your selling to everyone on es a week or a month later when I could have waited to get your bad ass charger
Sigma said:Alright good luck to you, would you happen to have an ETA of when yours will be done? lol I dont want to go spend $250 + about $100 shipping on that charger and then find out your selling to everyone on es a week or a month later when I could have waited to get your bad ass charger
Thanks
Well, I haven't got a ETA yet. My work is unpredictable and I use most of my free time on the charger - so I plan for it to be done at the end of summer. If I were you, I'd buy the charger and be happy - or aquire some PSU supplies and make your own boost converter (should be manageble if you're a fast learner).pelle242
100 W
just note that bms tacks on heafty shipping $100+....
IBScootn
100 W
Added to watchlist
hydro-one
10 kW
http://endless-sphere.com/forums/viewtopic.php?f=14&t=27199
u need two modded 63v meanwells which i can sell u for ....120$ .... that will get u too 126v...... enough?
u need two modded 63v meanwells which i can sell u for ....120$ .... that will get u too 126v...... enough?
