GGoodrum
1 MW
What is the charger voltage? It sounds like it is not a bit more than 3.68V x the number of channels.
-- Gary
-- Gary
Even Newer 4 to 24-cell Battery Management System (BMS)
- Thread starter GGoodrum
- Start date
steveo said:
All of the orange cell LEDs must be lit up before the auto shutoff can engage. You can test the control circuit by putting a 1k resistor between the ground and all shunts line. This should trip the main LED and make it stay green.
dermot
100 W
- Joined
- May 15, 2007
- Messages
- 172
I bought a couple of the version 2.2 boards last year, but events conspired against me so that I've not started on the A123 project until now.
Where is the most recent parts list/construction info so that I can get my v 2.2 boards working? I'm assuming that there have been several changes implemented in the interim.
dermot
Where is the most recent parts list/construction info so that I can get my v 2.2 boards working? I'm assuming that there have been several changes implemented in the interim.
dermot
Icewrench
10 kW
Just powered up my 3rd board and it had many bad things happening.
After a few hours of grrrrrrrr....
Turns out that a very close inspection showed some stray solder in several locations.
Cleaned up several small messes and made sure all traces and joints were good
checked clearance on all parts.
At least that`s what I think made it work. Good luck.
After a few hours of grrrrrrrr....
Turns out that a very close inspection showed some stray solder in several locations.
Cleaned up several small messes and made sure all traces and joints were good
checked clearance on all parts.
At least that`s what I think made it work. Good luck.
biohazardman
100 kW
dermot said:
PM me your e-mail addy and I will send you a PDF of the instructions and the BOM/parts list for the 2.2 board. Had mine together months ago but not the pack and bike. Just using it for my second time. These are nice even with my cheap SLA charger it works great.
Yes there have been changes in the board itself and some parts because of them.
Duane
Doctorbass
100 GW
pm_dawn said:
Hello and welcome pm_dawn,
I also know someone that want to use that bms in 40s configuration.
since the fet is a 100V, it need to be replaced ( 40s = higer than 100V !)
Fechter, what would you recommand to operate at 40s ?.. I may have a person that want to get 250+ of these 40s BMS so... :lol: finding the solution may help "a bit".. :wink:
Doc
pm_dawn said:
It sounds like either the all shunts or any shunts line is being held low.
You can test with a small variable power supply by feeding around 15v to the #4 cell tap wire. You could feed the first 8 with around 30v and do the same test. Use no cells for testing. With power applied you should see the main LED start out red and as you increase the voltage, eventually the shunt LEDs should light up and make the main LED go greenish. If the main LED goes green immediately and the cell LEDs are not lit, then I'd look for a short on the any shunts line. Look at the diagram I posted for Steveo. You can measure the voltage between the control ground and any shunts and all shunts lines. The control ground is connected to the charger negative wire, so you can attach one meter probe there if it is easier.
The any shunts line should be somewhere around 12v until the cell shunts come on full. The all shunts line should also be high and not go low until all of the shunts are on.
Keep in mind that if your previous test was done with no cells attached, it would be perfectly normal for the main LED to trip green immediately due to the auto shutoff feature. You can inhibit the auto shutoff by putting a jumper across any cell above cell 4.
To answer Doc's question, the charge control FET really only sees the voltage difference between the pack and the charger, so if the charger open circuit voltage is not too high, you may still be able to use a 4110 with 40s cells. You would need to measure the voltages. The 150v rated FETs have a much higher on resistance, so heating will be an issue if you go that route. It will work to use two or more FETs in parallel. A larger heat sink will help too.
What I would really recommend on higher power systems is to interface the BMS board to the charger circuit to do the throttling. This would require a modification to the charger, but it is not that hard in most cases. Inside the charger, there will be an optocoupler between the primary and secondary sides that controls the output. The idea is to make that opto come on whenever the any shunts line gets pulled low, causing the charger output to drop. This completely eliminates the FET and any heating/voltage issues there. The details on how to interface the charger depends on the charger design, so some reverse engineering would be needed. One possible way to do it would be to find the optocoupler inside the charger and place another opto in parallel with the existing one so that the outputs (mains side) are connected, but the input on the new one goes to the BMS board. The input side of this new optocoupler would get connected between the 12v line and the Any Shunts line (with a series resistor, around 1k).
Doctorbass
100 GW
fechter said:
Thanks you Fechter, I see now how it would work. The opto is a great idea.
I asked cause i wonder if the electrical insulation of the opto is ok for HV setup from the lowest cell to the highest of the entire serie.
Doc
steveo
100 kW
fechter said:
Hey Fechter
I have tried the testing with the 1k risistor and i haven't been able to have the bms to shut completly off.. All light light up ... And if you look at the main indicator led you can literly see it go from red to green .. however it remains as on orangish colour and doesn't not shut off the bms..
could you make further recommendations?
thanks
-steveo
steveo
100 kW
Hey Everyone,
I just wanted to ask what resistor needs to be changed on the bms board per channel to allow a 3.7v cut off for a123 batteries.
thanks
-steveo
I just wanted to ask what resistor needs to be changed on the bms board per channel to allow a 3.7v cut off for a123 batteries.
thanks
-steveo
steveo said:
Hmm.... try using a jumper wire instead of a 1k. Just don't hold it on there very long.
Measure the voltage on the all shunts line when all the LEDs are lit up too.
If it still won't stay tripped, then there would be a problem around the SCR part of the circuit. Make sure Q1 and D1 are installed in the right direction. You could measure the voltage on both sides of D1 against control ground. Breifly touch the jumper across All Shunts to ground then measure both sides again.
R101 and R102 control the voltage set point. The 75k/180k pair works out to around 3.7v usually. Is yours something different? You can change R101 to increase or decrease the voltage. There is a formula to calculate it.
Doctorbass said:
The ILD2 optocouplers are rated for 5,300 volts isolation. That should be enough.
One other consideration on large packs: The 12v regulator part may need to supply more current than the 78L12 can provide. It may be necessary to use a larger regulator. You could use a separate small 12v supply or use bigger parts on the existing on board regulator. Total current draw for 40s cells will be around 100ma.
steveo
100 kW
fechter said:
Hey Fechter
I have measured the ground and all shunt & shunt per channel .... all read .5v ...
I tried shorting ground and all shunt wire and yet no green light ...
Src & d1 are installed the right way .. and i used the diode setting in my multimeter to compare readings with another board .. and they where the same...
I could recall .. i think i switched the way d1 was installed at one point... could i have caused damage to any part when i installed in backwards?
I put a multimeter to the d1 .... and i get a reading both ways i put the multimeter... where as on working board .. i only get a reading in one direction...
the main led sit in an orangish red colour... and not pure red .. when its below the HVC mark ... something is odd here ..
-steveo
nieles
10 kW
is it possible to use this bms for charging a 10s LIPO pack?
steveo
100 kW
nieles said:
With modification to the HVC & lVC yes!!
-steveo
steveo said:
Yep, something is odd here.
I would try removing D1 and see if you can get the main LED to go solid red when its below the voltage set point.
With D1 removed, you could use the diode check.
Installing D1 backward should not blow anything.
.5v on the all shunts line is low enough to trigger the SCR.
You could first try measuring voltage from the control ground to both sides of D1. When the SCR trips, the side going to the SCR should pull low. When the board first comes on, that spot should be near 12v.
Hi,fechter said:
I ordered 2 boards and the components for my 38x3,2V (LiFePo4)cell-battery-pack and got some pretty good tipps from Gary about the higher voltage useage.
So I replaced the 6.8 ohm/5W shunt resistor with a 3.9 ohm/5W version in my order-list and will use a hetsink and/or fan.GGoodrum said:
That sounds good to me and I am looking forward to solder all togehter
Your post semms to me, that you think about something else to consider ?!?
You may be the first to run that many cells.
Like I posted earlier, I'm not sure about the 12v regulator. A separate 12v supply is one solution.
They are very inexpensive and not too hard to connect. It doesn't need to be very big, but I would suggest 0.5A output rating.
Like I posted earlier, I'm not sure about the 12v regulator. A separate 12v supply is one solution.
They are very inexpensive and not too hard to connect. It doesn't need to be very big, but I would suggest 0.5A output rating.
steveo
100 kW
fechter said:
Hey Fechter,
I have measured the control ground & both side of the d1 diode.. The top side of the diode measures 1.4v ... the bottom of the diode measured .75v - this done with the bms below the HVC mark
With the Bms at the HCV point .. i measure .53v on the top of d1 .. and .74v at the bottom of d1
with d1 removed ....the light still stays orange below the HVC mark
thanks
-steveo
OK. I surely will ask you some questions, when I got all the components and solder all together.fechter said:
Right at the moment, I lost a little overview :? but if I smell the first solder-smoke, I take a closer look. thanks so far
after looking at the schematic i came up with another idea.
i was wondering since the las 4 cells are the last to be charged if i where to short the all shunts active connect to ground resulting in a temporary disabling of the auto shut off and let it that way overnight would that make the 4 cells come up and balance out better?
the theory is the reason that the first 12 cells stay fully charged (if charge cycle is restarted hours after charge stops for example) is because they are being held at the charge level until the other 4 finish.
another way that should work and would be safer for the bms ( shorting somethings to ground could fry electronics and i am not sure if applying a logic low to all shunts active by connecting to ground is safe) would be to discharge the other 12 cells by putting a load on them so the other 4 cells would end up being fully charged first.
a 3rd option would be to connect a car battery charger to the 4 cells to bring them up to partial charge then swap chargers and use the normal charger to charge the pack so the 4 cells would come up first and hold there until the rest charges.
correct me if i am wrong but i remember the bms was originally designed for a123 cells witch charges slightly differently so the board works perfectly with them.
i am using the 40138 cells that andy h sells witch may have a slightly different charging specs that may cause the problem on the 4 cells.
the charger is the soneil 4808 ( i believe) i bought from batterymart http://www.batterymart.com/p-battery-defender-48v-3_5a-battery-charger.html battery mart claims 3.5 a witch is probably correct even though the charger claims 10 a.
the charger had to be set to 59.2 volts and that was for the original pack made from 2 28v dewalt nano packs but do i need to change it for the 40138 cells?
i was wondering since the las 4 cells are the last to be charged if i where to short the all shunts active connect to ground resulting in a temporary disabling of the auto shut off and let it that way overnight would that make the 4 cells come up and balance out better?
the theory is the reason that the first 12 cells stay fully charged (if charge cycle is restarted hours after charge stops for example) is because they are being held at the charge level until the other 4 finish.
another way that should work and would be safer for the bms ( shorting somethings to ground could fry electronics and i am not sure if applying a logic low to all shunts active by connecting to ground is safe) would be to discharge the other 12 cells by putting a load on them so the other 4 cells would end up being fully charged first.
a 3rd option would be to connect a car battery charger to the 4 cells to bring them up to partial charge then swap chargers and use the normal charger to charge the pack so the 4 cells would come up first and hold there until the rest charges.
correct me if i am wrong but i remember the bms was originally designed for a123 cells witch charges slightly differently so the board works perfectly with them.
i am using the 40138 cells that andy h sells witch may have a slightly different charging specs that may cause the problem on the 4 cells.
the charger is the soneil 4808 ( i believe) i bought from batterymart http://www.batterymart.com/p-battery-defender-48v-3_5a-battery-charger.html battery mart claims 3.5 a witch is probably correct even though the charger claims 10 a.
the charger had to be set to 59.2 volts and that was for the original pack made from 2 28v dewalt nano packs but do i need to change it for the 40138 cells?
ejonesss said:
AndyH
10 kW
Hi Eric,
Which version BMS did you build?
59.2V is 3.7V per cell. It might not quite be enough voltge to get the BMS to fully balance and latch off. You may need to adjust up to 59.7V.
The PSI cells you're using are standard LiFePO4 and are a perfect match to the BMS. You don't have to modify the BMS in any way to charge and balance the pack, provided the BMS is working correctly.
The surface charge is just that - a surface charge. There's very, very little energy stored in that area around 3.7V. The charge curve is so steep there. It's like adding individual drops to refill a Coke bottle - it takes forever to notice the liquid level rising until it reaches the neck of the bottle where it starts to narrow - then it looks like each drop makes the level jump. In other words - a very, very small movement of power makes the voltge move a great deal. Don't get too hung up on taking charge points to four decimal places.
Andy
Which version BMS did you build?
59.2V is 3.7V per cell. It might not quite be enough voltge to get the BMS to fully balance and latch off. You may need to adjust up to 59.7V.
The PSI cells you're using are standard LiFePO4 and are a perfect match to the BMS. You don't have to modify the BMS in any way to charge and balance the pack, provided the BMS is working correctly.
The surface charge is just that - a surface charge. There's very, very little energy stored in that area around 3.7V. The charge curve is so steep there. It's like adding individual drops to refill a Coke bottle - it takes forever to notice the liquid level rising until it reaches the neck of the bottle where it starts to narrow - then it looks like each drop makes the level jump. In other words - a very, very small movement of power makes the voltge move a great deal. Don't get too hung up on taking charge points to four decimal places.
Andy
version 2.2
also the issue is that if for example i put the battery on to charge after i am done using it for the day and leave it on overnight.
it finishes charging later at night and sits green and by morning if i unplug and reconnect the charger the 4 cells will start charging as if a load had been placed on them.
that is what makes me believe that the bms may not be holding the voltage long enough.
that is why my theory of partially charging the 4 cells with a car charger or majorly discharging one of the others to cause the 4 to be held until the others finish should work.
if the effect can be reversed and cause the + end cell(s) to act that way then it may be that the bms is not holding the voltage long enough to teach the cells that that is the full point.
i know this because when i reconnect the charger in the morning 12 leds light up immediately while the 4 done .
i will pm you with the other info since it pertains more to your store purchases.
also the issue is that if for example i put the battery on to charge after i am done using it for the day and leave it on overnight.
it finishes charging later at night and sits green and by morning if i unplug and reconnect the charger the 4 cells will start charging as if a load had been placed on them.
that is what makes me believe that the bms may not be holding the voltage long enough.
that is why my theory of partially charging the 4 cells with a car charger or majorly discharging one of the others to cause the 4 to be held until the others finish should work.
if the effect can be reversed and cause the + end cell(s) to act that way then it may be that the bms is not holding the voltage long enough to teach the cells that that is the full point.
i know this because when i reconnect the charger in the morning 12 leds light up immediately while the 4 done .
i will pm you with the other info since it pertains more to your store purchases.
AndyH said:
ComponentKiller
1 µW
- Joined
- Aug 26, 2009
- Messages
- 1
Hello all of you,
This is my first post at this amazing forum, after many hours of reading from the side. It is a pleasure to see the large activity in the DIY EV world. There is not that much going on it the area here in Denmark yet, so forums as information rich as this one are gold mines. Thanks to all the great contributors.
Now to my question, which I hope you can help me solving:
I have assembled a 16 channel version 2.5 BMS and tested the LVC and HVC functions for each channel. They all work as expected. The same is the case, when I connect it to my 70Ah Sky Energy cells and connect my charger. The charging runs fine - both at 4A and 18A.
The problem arises when the cells start to get full. I will try to describe my observations. I hope it will not get too complex, even though these matters are not easyly described in writing.
The diodes are blinking at a rather low frequency - something like 10Hz or less. I have not measured it, but it is visible to the eye (I had expected it to be faster). The main LED turns almost green, when the shunting starts.
In the beginning, the charger turned off shortly after the shunting started on just a few of the channels. It was putting out 59.2V, but apparently the slow shunting made it see the closed FET primarily, which made it turn up the voltage to the max and turn off, before the rest of the cells could keep up.
This made me open my 4A charger (from http://www.BMSbattery.com. I have two chargers from there, which - by the way - seem to have exactly the advertised output all the way up to the max voltage), and set the max voltage higher. This has helped a little, and with many charger resets, I have got to the point where 13 or 14 of the cells are fully charged and only a couple are lagging behind.
After I have rised the charging voltage all the way up to the mid 60'es and only a few cells are not fully charged, the BMS is cutting of the charging and the main LED goes green after around 30 seconds of charging. Then the channel diodes turns off after a few seconds - except for the first three channels, which stays lit for slightly longer than the rest. Channel number one seems to be the last one to turn off each time in these cases - unless I pull the power to the charger (still leaving it connected to the BMS so the charging logic is powered). In that case the diodes of the first channels turns off simultaneously with the charger(?).
When the diodes turn off, the main LED turns slightly red'ish again, the charger resumes charging and the shunting starts again on the fully charged cells. This goes on for around 30 seconds, before the BMS again cuts off the charging with a green main LED, and so on.
If this kept going, I presume the last cells would ultimately catch up with the full ones. The charger does, however, switch to the "The battery is fully charged"-state and stops charging just after the charging starts again in the second or third cycle. I presume this is due to a function in the charger that makes it stop charging when the load has not allowed any current into it for a while (as a full battery would).
So, as you see, this leaves me wondering:
What is the shunting frequency supposed to be?
Why does the BMS cut of the charging before the cells are full?
Why do the first cells take so long time to discharge back to HVC when the charger is on, and not when the charger is off?
Another thing I wonder about is that the "All shunts active"-line seems to creep towards somewhere around 0.7V when 13-14 of the shunts are active. There still seems to be a 12V drop over the SCR, though, so I presume it has not been triggered. As expected, when some cells are not fully charged, the main LED never stays green, after the last channels have stopped shunting.
I hope you have some clever inputs on my problems, so I can get this device to work as great as expected. I am going to used the battery in my sailing boat along with a brushed Mars ME0708.
-Niels
This is my first post at this amazing forum, after many hours of reading from the side. It is a pleasure to see the large activity in the DIY EV world. There is not that much going on it the area here in Denmark yet, so forums as information rich as this one are gold mines. Thanks to all the great contributors.
Now to my question, which I hope you can help me solving:
I have assembled a 16 channel version 2.5 BMS and tested the LVC and HVC functions for each channel. They all work as expected. The same is the case, when I connect it to my 70Ah Sky Energy cells and connect my charger. The charging runs fine - both at 4A and 18A.
The problem arises when the cells start to get full. I will try to describe my observations. I hope it will not get too complex, even though these matters are not easyly described in writing.
The diodes are blinking at a rather low frequency - something like 10Hz or less. I have not measured it, but it is visible to the eye (I had expected it to be faster). The main LED turns almost green, when the shunting starts.
In the beginning, the charger turned off shortly after the shunting started on just a few of the channels. It was putting out 59.2V, but apparently the slow shunting made it see the closed FET primarily, which made it turn up the voltage to the max and turn off, before the rest of the cells could keep up.
This made me open my 4A charger (from http://www.BMSbattery.com. I have two chargers from there, which - by the way - seem to have exactly the advertised output all the way up to the max voltage), and set the max voltage higher. This has helped a little, and with many charger resets, I have got to the point where 13 or 14 of the cells are fully charged and only a couple are lagging behind.
After I have rised the charging voltage all the way up to the mid 60'es and only a few cells are not fully charged, the BMS is cutting of the charging and the main LED goes green after around 30 seconds of charging. Then the channel diodes turns off after a few seconds - except for the first three channels, which stays lit for slightly longer than the rest. Channel number one seems to be the last one to turn off each time in these cases - unless I pull the power to the charger (still leaving it connected to the BMS so the charging logic is powered). In that case the diodes of the first channels turns off simultaneously with the charger(?).
When the diodes turn off, the main LED turns slightly red'ish again, the charger resumes charging and the shunting starts again on the fully charged cells. This goes on for around 30 seconds, before the BMS again cuts off the charging with a green main LED, and so on.
If this kept going, I presume the last cells would ultimately catch up with the full ones. The charger does, however, switch to the "The battery is fully charged"-state and stops charging just after the charging starts again in the second or third cycle. I presume this is due to a function in the charger that makes it stop charging when the load has not allowed any current into it for a while (as a full battery would).
So, as you see, this leaves me wondering:
What is the shunting frequency supposed to be?
Why does the BMS cut of the charging before the cells are full?
Why do the first cells take so long time to discharge back to HVC when the charger is on, and not when the charger is off?
Another thing I wonder about is that the "All shunts active"-line seems to creep towards somewhere around 0.7V when 13-14 of the shunts are active. There still seems to be a 12V drop over the SCR, though, so I presume it has not been triggered. As expected, when some cells are not fully charged, the main LED never stays green, after the last channels have stopped shunting.
I hope you have some clever inputs on my problems, so I can get this device to work as great as expected. I am going to used the battery in my sailing boat along with a brushed Mars ME0708.
-Niels
Similar threads
