Dewalt A123 BMS battery interface solution

[nightowlgk]

Sorry, I only have Dewalts to play with. No plans on using any other packs since the Dewalt have been very reliable. I will be getting the Dewalt 28volt packs, and will verify if those will work.

Yeah, but at $50 each!!!??? Bosch are much more attractive for multi setups.

[kfong]

I've found a problem with some of the Dewalt batteries. Some of the older Dewalts when enabled have a rather high discharge current when in standby mode. This is not problem when the pack is charged since it takes a very long time to drain a pack when full, but when the pack is empty and left plugged in, it will fully discharge due to small constant drain on the batteries. I don't recommend leaving the packs installed after discharged because of this, unless you verify you have one of the newer bms designs. On my setup, I usually take the pack out to charge after a run and have them in my charger ready for my next ride, so this isn't a problem I need to worry about. For those of you who plan to have the batteries permanently part of the bike, you will need to disable the enable line and add an opto circuit to turn them on. I'm working on an opto board to do this and should have the boards sometime next week.

[Tony]

I'm going to be using the Dewalt 28v packs with your boards, would the same resistors and diode be used as the 36v pack?

[kfong]

I have the 28volt packs to play with now. Seems to work, will be doing further testing. Been busy lately with other projects, but I did plug it in and was able to see the proper voltage reading at least. The only change would be the resistor values, but I suspect that won't even be necessary since it's probably setup for 20amps, similar to the 36volt version.

[Evereste38]

Hello,

I'm try to use the 28V pack too, but I'm not able to activate the BMS with the 3 resistors like on the 36V pack.
I blew the fuse and shunt it by a external between the BAT- inside and the output Pin outside, ratter to dig into the pack to change the SMT fuse, does it a problem ?
Do you confirm your pack in 28V is working with the same resistor network ?

Thanks.

[webfootguy]

Did you blow the fuse before or after trying to use the interface board? I just got the rest of my parts to build 2 interface boards today and I have lots of 28v packs so I will give it a try fairly shortly.

[Tony]

* Report this post
* Reply with quote

Re: Dewalt A123 BMS battery interface solution

Postby kfong » Sat Jun 20, 2009 10:02 am
I did some more testing with the boards. Looks like capacitive spikes are a problem. It will shut down the BMS, an inline inductor of 300uh and a diode across the batteries will be needed. This is what was in the original circuit. I was hoping to avoid using them. I will have to do more testing.
Last edited by kfong on Sat Jun 20, 2009 8:02 pm, edited 1 time in total.
1st ebike (Diamond Back, 650w cylcone, 3 dewalt packs)
viewto ... f=3&t=6348
2nd ebike (Motobecane DS, 600w BMC torque, 10S8P Konions)
viewtopic.php?f=6&t=12654
Outback Mountainboard
viewtopic.php?f=12&t=11137
X-Treme Electric Scooter

Dewalt interface board
viewto ... 14&t=10986

My website http://www.embeddedtronics.com/
RC blog http://www.rcgroups.com/forums/member.php?u=23712

kfong
100 W
100 W

Posts: 177
Joined: Tue Sep 02, 2008 10:34 am

* Private message
* Website

Have you tried this? maybe it will prevent the BMS from shutting down.I'm waiting for all my parts to arrive too for the 28v set up, should be testing soon-

[kfong]

I finally got around to testing the 28volt packs last night. They work fine with the interface board. Here is the chart of the cutoff, as you can see the BMS shuts down at around 21volts. I put one pack on the bike this morning to do a current test. The max current I was able to pull was close to 25amps. This is actually a higher setting than the 36volt setup. This was a pleasant surprise. I verified this with a second pack. Note: you need the inductor to keep the BMS from shutting down due to the current spikes. I’m guessing since the voltage is lower, Dewalt increased the current limit to make up for the power.

[kfong]

Doesn't sound like you are using my boards since I just verified that it works with the 28volt packs.

Hello,

I'm try to use the 28V pack too, but I'm not able to activate the BMS with the 3 resistors like on the 36V pack.
I blew the fuse and shunt it by a external between the BAT- inside and the output Pin outside, ratter to dig into the pack to change the SMT fuse, does it a problem ?
Do you confirm your pack in 28V is working with the same resistor network ?

Thanks.

[webfootguy]

kfong, I'm about to finishing building mine and test them as well on the 28v packs. What inductor do you recommend and where should it be inserted in the circuit? I'm assuming either the + or - pack would be ideal. Does the inductor have to be rated at +25 amps? Those seem spendy. Can you suggest a part to experiment with?

[Evereste38]

Did you blow the fuse before or after trying to use the interface board? I just got the rest of my parts to build 2 interface boards today and I have lots of 28v packs so I will give it a try fairly shortly.

I blow the fuse before to find this forum and the board of Kfong. So don't worry Kfong made a pretty good job and you can trust his tests.

Kfong:
Yes, I don't use your board yet, I just tried quickly with 3 resistors to be sure that my pack is still working.
What do you think, please ?
Does my fix for the fuse bad ?
Do I have to buy another pack to use your activation board ?

For remember and more detail: I shunt the fuse inside the BMS by a polyswitch resetable one of 2.5A beetween the battery - polarity and the extarnal pin 1.

[kfong]

The easiest thing to do is find a toroid shaped donut. See earlier pictures and just wind heavy gauge wire around it. A transformer iron core would probably work as well, if you can get one torn apart and just use the rectangle portion of it. Digikey sells a variety of ferrite cores. Under ferrite inductors in their search engine. Just chose one that you can wrap some heavy wire through. I just kept wrapping as much wire as I could and it worked fine. I really didn’t try to optimize it. I'm sure I could of used less wire, but the higher the inductance the more it prevents spikes. Value isn't too critical, it will either work or you will need more winds or a bigger ferrite. The ferrite core goes between the batteries and the controller on the positive side. I wrapped the positive wire around the toroid as you can see the red wires in the earlier post. The thick wires should be the same gauge as the existing ones used for the power leads. Works amazingly well.

Just go by package size
http://search.digikey.com/scripts/DkSea ... 0%20toroid

For reference the dimension of the ferrite toroid I used was 47mm in diam. 18mm thick with a 24mm hole. It was just something I found in my pile of junk electronics. You just need something big enough to be able to wind the thick wires around to create a large enough inductor.

kfong, I'm about to finishing building mine and test them as well on the 28v packs. What inductor do you recommend and where should it be inserted in the circuit? I'm assuming either the + or - pack would be ideal. Does the inductor have to be rated at +25 amps? Those seem spendy. Can you suggest a part to experiment with?

[kfong]

As long as you fixed the fuse, you should have no problem using my boards. It will still work with the fuse blown since the power is taken from the fet output. The Dewalt charger might need the fuse fixed to charge the battery though. That fuse is 15 amps, but since you won't be using it for power. It only needs to be high enough for the charging current.

Did you blow the fuse before or after trying to use the interface board? I just got the rest of my parts to build 2 interface boards today and I have lots of 28v packs so I will give it a try fairly shortly.

I blow the fuse before to find this forum and the board of Kfong. So don't worry Kfong made a pretty good job and you can trust his tests.

Kfong:
Yes, I don't use your board yet, I just tried quickly with 3 resistors to be sure that my pack is still working.
What do you think, please ?
Does my fix for the fuse bad ?
Do I have to buy another pack to use your activation board ?

For remember and more detail: I shunt the fuse inside the BMS by a polyswitch resetable one of 2.5A beetween the battery - polarity and the extarnal pin 1.

[Evereste38]

[quote="kfong"]As long as you fixed the fuse, you should have no problem using my boards. It will still work with the fuse blown since the power is taken from the fet output. The Dewalt charger might need the fuse fixed to charge the battery though.

Thank so much for this very quick response.

[sschoon86]

Hi i am fairly uneducated in electronic terminology so this may seem like an obvious question to many. The schottky diode is on backorder and it looks like they are anticipating at least a month out. Is there any way I could replace it with a V30100C-E3/4W? This appears to be almost exactly the same other than the surge amps is reduced to 160A instead of 250A. Digikey only sells these in multiple of 1000 but mouser sells them individually and the have a whole grundle of them available. Can you help me.

[kfong]

Should be fine.
Kin

Hi i am fairly uneducated in electronic terminology so this may seem like an obvious question to many. The schottky diode is on backorder and it looks like they are anticipating at least a month out. Is there any way I could replace it with a V30100C-E3/4W? This appears to be almost exactly the same other than the surge amps is reduced to 160A instead of 250A. Digikey only sells these in multiple of 1000 but mouser sells them individually and the have a whole grundle of them available. Can you help me.

[kfong]

Recieved this email from Jeff, who did the original design. Useful info.

Hi Kin,

Very nice implementation of my original circuit for discharging the Dewalt packs. I've upgraded this circuit for use with a 2 Hp motor and 600 uf filter capacitor. The only necessary change was to double the size of the series inductor, further limiting inrush current. Interesting that you found two flavors of pack output circuitry. Overcurrent trip may depend on some subtle difference, like conductor gauge or FET lot. To be safe, I would recommend keeping the inductor on all packs, even if they don't appear to need them.

Please check out my recent thread on Dewalt, A123 Nanophosphate packs disappearing from market. How do the Konions compare to Dewalt packs?

Best Regards,
Jeff Radtke
(CyclemotorEngineer)CyclemotorEngineer
-------------------------------------------------------
- 406/409 on 16 inch wheel
- 72 Volt, 20 A controller
- 2s2p unmodified DeWalt DC9360 packs
- 1 hour charge, w/ (4) DeWalt DC9000
- 3 inch travel, trailing link suspension
- 30 second Cyclemotor installation or removal
- 27 MPH maximum assisted speed
- 38 MPH downhill speed test
-------------------------------------------------------

[methods]

Hmmm... Much Respect

Member told me about this thread over the phone.
Glad I got to see it, very cool.
Quality PCB's too - bet those were expensive.
recoup your investment yet ?

If there is one thing we are lacking around here it is plug and play BMS / Charging / packaging solutions
This would be something I could point a noob to and feel confident that they would succeed.

Keep up the good work

-methods

[CyclemotorEngineer]

Just to update regarding the effectiveness of using an inductor to prevent BMS trip:

I learned that the Dewalt BMS would trip from the 72 volt, 404 motor revving up after hitting a bump. When the wheel hit the ground again, the instantaneous load would peak, even higher than when upon initial power-up. Increasing the 20A controller capacitance by 470uF and doubling discharge series inductance solved this perfectly.

There was a discussion some time ago about using series inductance in this discharge circuit, which concluded with many suggesting it was unwise. Seems that people were using a higher power motor that my original tests, or refusing to use or increase size of the series inductor. Arguments were based on theoretical grounds by people who had not done the experiments. What really happens in the BMS during transients occurs on short enough time scale that the 600 uH inductance has a significant effect. I have done the math, but the experiment is what really counts, since the Dewalt BMS remains a black or grey box at best. I still maintain that this is a great solution, since we have DC in discharge circuit, unless if controller filter capacitor is undersized, or transients are encountered during startup or on rough roads. Total power dissipated in the inductor during these transients is tiny. Weight of inductor is negligible, and it is operationally transparent.

This plug and play solution is ideal for anyone wishing to use Dewalt packs, unless they want to discharge more than 10 to 15A per pack, in which case the lifetime of the cells are reduced anyway. Nothing wrong with Dewalt BMS, if you use it with the DC9000 charger. The DC9280 uses same charger and discharge connector as the DC9360. Dewalt DC9280s are running 80 cents per Watt-hour at the moment. This is a better price than the Konions. Trouble is that price will probably not last long because of resolution of a patent dispute. (See other recent thread regarding A123's exit from the power tool market.) Hopefully there will be something better soon, but anybody who knows is probably paid to remain mute.

[kfong]

Thanks Methods for the compliments and CyclemotorEngineer for the update. Someone had asked if I could post a picture of the battery carrier. Here is what it looks like in PDF. I am coming close to 2000 miles on my packs without any issues. They have been very consistent and have not shown any loss in amp hours. I run 3 packs at a time when riding through the trails and swap them out for a second run. It usually takes me an hour and a half of nonstop trail riding. Rode today, but the weather is starting to get crappy here in Michigan so I have to pick my days. I’ve been very happy with the A123’s. This is the second season and I’m sure it will have no trouble lasting the stated cycle life.

[kfong]

I've been asked to post some more details on how I mounted the interface boards.

Here are pictures of the battery tray I built. I used lexan for the carrier and a lexan spacer between the fiberglass board. The lexan spacer was needed to provide a gap for the screw heads, I would omit it if I could. Wood or aluminum channel could be used. It doesn't have to be as elaborate as my setup. The interface board is epoxied while plugged into a pack so the glue can set while in final position. Once done, the batteries just plug in and lock into position.

[Lyen]

I love your solution! In addition, I have figured another way to save space while finding a good place to put and hold my batteries. This way, I can transport the packs from one bike to another with whatever is in the bag easily. I use it for commute 10 miles to work almost everyday except in the rain. See below:

[kfong]

Lyen,
Where did you get the Dewalt battery holders. What were they originally used for? They look like a good mounting solution.

[kfong]

Is there any interest in complete built boards. I'm thinking of charging $35 for them.

[Lyen]

Lyen,
Where did you get the Dewalt battery holders. What were they originally used for? They look like a good mounting solution.

They are simply the main part of flashlights sans the gooseneck & the bulb pieces. Can't you recognize that? LOL