Headway Pack Builders thread

use sections cut from the sides of gallon size plastic milk jugs. you can tack it on the inside of the metal with contact adhesive, or liquid nail construction adhesive. looks nice though. pretty smooth curves.
 
dnmun said:
use sections cut from the sides of gallon size plastic milk jugs. you can tack it on the inside of the metal with contact adhesive, or liquid nail construction adhesive. looks nice though. pretty smooth curves.
Click to expand...

Thanks, and good suggestion. I'm bummed because I'd have double my range if ev-components stood by their product. The ph. no. on their page doesn't work, nor did they respond to their email form or direct email.

One lousy defective cell holding me up.

000t734w.jpg


Looks like I'm switching to lithiumdepot.
 
flügelwagen said:
I'm bummed because I'd have double my range if ev-components stood by their product. The ph. no. on their page doesn't work, nor did they respond to their email form or direct email.
Click to expand...
You might wanna search this forum and DIY Electric Car forums for why they aren't responding...seems that their fearful leader has done unright things. :(
 
Gee whiz- did I kill this thread? Have y'all moved on from Headways? Wouldn't fault you if you did. I no longer have any confidence in running these without a BMS. As often is the case, I wanted to believe that the easy way was the righteous path, but was stricken down by reality in the form of hissing, dripping Headways. So after retiring my casualties, I bought a Signalab 2.5 (presumably) BMS and installed it in the oil-tank area. Because the BMS did not come with any documentation I used this older version circuit diagram Headway Headquarters displays from all-battery.com;
12c.JPG
I dreaded the complexity of the loom, but it really wasn't too hard. I studied dnmun's excellent post earlier in this thread and got soldering. I had to lengthen most of the sense wires and by punching two slots in each buss-bar with a chisel, found a way to crimp and solder them in a very secure way.
s640x480

The harder task was how to mount this naked circuit-board in the bike. I couldn't find nylon standoffs anywhere, and didn't like the available enclosures, so I came up with an Earth-friendly solution- chop up a DVD case;
s640x480
It is slim, insulating, and transparent so I can see the leds. The first cycle I got only one light to go on and my 43v charger was cut-off at 40v. After consulting the ES oracle I knew I had to drain the outlier, high cell and cycle again. Here is a pic of that, using a signal-lightbulb and some telephone wire;
s640x480
You can see I had my charger on deck to top up low cells, but I was not sure the sense wires were fat enough to handle the current and there was no way I was going to fry my new loom, so I didn't risk it. After a few rides and one more scene like this with the light-bulb, I am now up to 42v and 4 leds, so confidence is high. So if anyone is still watching the Headway Pack Builder's Thread, Flügelwagen is going on record as a BMS convert, at least when it comes to Headways.
 
i used the v 2.5 signalab on my 12S headways originally so i could watch the leds, but now i use a v1 signalab because the shunt current is higher.

sorry you lost some cells without the BMS originally, hope others can learn from this too. i use a fat ceramic power resistor,1 ohm, 25W to drain the high cells when balancing.

be careful not to assume the BMS will actually shut down the output if any cell drops below 2V. i have had cells drop to that level while doing capacity tests, even though the pack later shut itself down while powering the watt meter for a long time, so the LVC was actually working, but not observed to shut off at 2.1V. just FYI.
 
Yea, I'm struggling with my own situation of bms, headways, and balancing. In a rush to get the bike to the Undead Race in Tucson, I hooked up the first of two bms system wrong. I had no wiring diagram and looked for a clue online and guessed wrong. The bms burned no. 1 resister. The second bms I hooked up correctly after seeing what had happened to the first. This made me one bms short for my setup and it was early morning before the race and no hope of parts to repair it. I used the race bike at the race (in practice) before the race, and fully intended to race it until I got a flat tire and had to switch to my daily rider. I took the chance that all my cells would discharge evenly because they had been previously broken in. Did not happen that way. I killed 3 cells. Many hours of repairing the packs and I only have one single cell charger. Between trying to lower the new high cells and raising the low cells it has been a chore. The bms really does not want to handle a wide range of difference in cell voltage. I had purchased a version 2.5 Signalab from Hightekbikes but it looks so small in comparison to the bms that I toasted, that I questioned Terry if it could hold up to the amps. Since I am still unsure, I am trying to fix my smoked bms. I replaced the melted resistor with one that is supposed to have the same rating but I am skeptical. If I jumper across where the charger negative contact is on the board, to the first fet (I believe the drain middle leg)), the charger appears to charge the pack. What does this tell me? That the resister that I replaced is tripping the no charge state or is the fet bad? Hmmmm.... I think the jumper bipasses the board. I still suspect it has to do with an incorrect resistor on number one...Help!P1010235 (600 x 450).jpgLook at the size difference of these bms's.P1010237 (600 x 450).jpg
 
Not sure of the cause, but I am pretty sure that I have those exact same resistors on one of the older small Signalab BMS units I have here. You're welcome to a couple of the resistors if you want to either meet up or I can mail them to you if you PM me an address, after I verify if theyr'e the same or not. Gotta go check, won't be until tomorrow night or next day.

The main differences in the size between the two BMSs:
--Shunt resistors are much larger and capable of dissipating much higher power (and thus balance faster) on the larger one.
--Discharge FETs, ditto. (and better heatsinked)
 
The painless way to balance is with a series resistance to limit the charging current to what the BMS shunts can handle. Then you can forget about it while the BMS does its thing. At 100 milliamps it would take 50 hours to correct a 5 amp-hour difference.

I use Andersons which makes it easy to patch in a cable having banana plugs for a current meter and a string of 5 ohm resistors and an alligator on a pigtail to choose the operating current. If the charger wants to shut off at that current you have to add a load in parallel to keep it running.
 
Thanks for the response you guys. I'll get with you on the PM side Amberwolf. Dak, I sure would enjoy pictures of what you are describing, if you can. I am sure it could help others who venture into this territory also. I am also playing with dnmun on the PM also. All the help is appreciated.... thanks again!
 
Hay Evoforce, Signalab also makes a 60 amp. v2.5 for a 48v and the white 16 pin will plug right in to a 40amp , but the bms side looks the same ? Maybe signalab has a higher amp. 36v model. Ping has signalab in one of his email address is Ping signalab himself ? One and the same ?
 
Thanks 99 yes mine is a 12s setup doubled (36x2) to make 72 volts. Maybe I can return this small Signalab bms to Hightekbikes.
 
Heck, if all that's needed is a higher discharge current capability in any BMS like these, you just add more parallel FETs or replace the existing ones with better ones (if there are any), and heatsink them much better, perhaps with active cooling. :)

But it ought to be possible to easily fix the damaged BMS, simply by comparing (in operation) the still-good one to it, and seeing what signals are doing what, in each component related to that damaged channel and in the charge-enable circuit. One of them is at fault, and I doubt that it is the resistor (but it is possible, I suppose).
 
I have two non-working Ping-type BMS that have those types of resistors, I just have to get the box out and see if they are the same value yours are. I"m guessing yours are 11ohm based on this pic:
http://endless-sphere.com/forums/download/file.php?id=51454
that makes the stripes look fairly brown, but they could be 22ohm. Either way, looks like gold band next, 5%; I forget what the fifth band means but I don't recall a fifth on my resistors. Same physical size, though, AFAICT.

I'm pretty sure one of them is not, as I think it was 22ohm, 5%, 2-3W. That was a Victoria or Victoriz BMS that looks exactly like a Signalab BMS, from Karma (it's already got 3 of the resistors removed, presumably to fix other things he was working on).

The other is an actual Signalab, but I haven't seen it in long enough that I can't recall the resistor bands. :( So I need to dig it out.

I also have a lot of other power resistors salvaged from many things, and one of them might be the one you need (or very close).
 
dnmun said:
i used the v 2.5 signalab on my 12S headways originally so i could watch the leds, but now i use a v1 signalab because the shunt current is higher.
Click to expand...

Hmm, good to know. The more I watch these red lights, the more I wonder how useful they are. It seems that the useful info I'm getting is from the 12-pin female plug.

dnmun said:
be careful not to assume the BMS will actually shut down the output if any cell drops below 2V. i have had cells drop to that level while doing capacity tests, even though the pack later shut itself down while powering the watt meter for a long time, so the LVC was actually working, but not observed to shut off at 2.1V. just FYI.
Click to expand...

Well, that has not yet been my experience. Mine starts cutting off power by pulsing the current long before my controller's LVC kicks in, and so far it has not let a cell get below 2.3v.
 
Hello,

Following to a bad experience with a capacitor BMS (model : AE-LMD17 REV A1) : http://endless-sphere.com/forums/viewtopic.php?f=14&t=24646 ;
I'm looking for a reliable solution to control my 16 Headway 40160s cells.

For the moment I found this solutions :

- The Signalab BMS v2.5, 16cells 30A (who's the best seller for this model ?)
- The GGoodrum & Fetcher BMS (I'm very interested by this model but still under development and unavailable at this time ?)
- The 16cells 40A PCM from batteryspace (a bit expensive with shipping rates to Europe)
- A "Headway BMS" that I don't know where could it be bought.
- ... Any other solution ?

I need a BMS able to deliver around 30/40A continous in discharge, 10A in charge with LHC, HVC and balancing (during charge/discharge) functions.

The GGoodrum&Fetcher BMS seems to me to be the best solution but I don't know when it will be available again...

Thanks to give me your opinion and your advices on this products or on other more reliable solutions.

PY
 
All,

We are building a 4P24S pack for an electric motorcycle project using 15Ah 40160S cells. I'm seeking input on the best method of making the parallel busbar connections: hard copper busbar, polyswitch, or fusable link.

I see most people on this thread using a hard copper busbar. We favor this approach because it is easy and it allows the most compact configuration. However it concerns us for a couple of reasons. First, vibrations are transmitted across all the cells in the block. We'll be using Headway connector blocks but I'm sure there will still be some motion at the terminals. Second, if a cell fails as a dead short, then all hell breaks loose. I have read some reports on ES concerned about cells failing this way, but does it really happen?

If necessary, we could make the 4P connection with polyswitches - at a considerable hassle in pack fabrication. The resettable fuses themselves aren't very expensive (about $0.50 in quantity) but getting polyswitches with a rating over 30V seems to be a challenge (at least I can't find them on Mouser).

Finally, we see some builders using small guage wire (maybe 16G) intended to serve as a fusable link.

Any advice from the community on the the best practice for parallel connections would be helpful. A few constraints on our project: (a) it will be a motorcycle with all the accompanying vibrations and (b) we are building a consumer product so we need to be able to sleep at night knowing that battery packs out of our control are being charged and discharged by users with little knowledge of Li-ion hazards.

Thanks in advance for any help!

Tom McKinnon
 
this has been a point of contention already. it is clear to me that current flows serially through the individual cells from one end to the other, and does not flow side to side through the parallel connections during discharge.

therefore you should connect the individual cells in series with large gauge wire and the parallel connection can be of very small gauge wire capable of handling the shunt balancing current only. you should not be using polyfuses imo, and the terminal connection at the positive and negative ends of the pack should be of sufficient capacity to handle the total current expected from all of the series connections added together and collected into the buss bar. to compensate for voltage drops across the buss bar the terminal lug or lugs should mounted on the buss bar in a symmetrical pattern, imo.
 
I will probably buy soon an "Headway" BMS to finalize my EV moped project but I've a last question about the balancing current...
I've a 48v 16Ah battery pack (16S1P Headway 40160S) that I charge with a 600W KP series charger (58.4V / 10A) (For the moment I balance my cells
manually for my tests).
But I'm wondering if a 75mA balancing current will be enough to balance my pack with a 10A charger ? My charger will blink many times at the end of the charge before the end of the balancing...
 
If some cells are low and the charger bangs out an amp the bypasses on the other cells will not handle it so one will overvoltage rather quickly which will trigger the BMS to disconnect so the charger cuts off. The BMS bleeds the high cells down over a few seconds, the charger kicks in again, rinse and repeat. Not good, but eventually the low cells will come up.

You could prevent that from happening with a series resistor that limits the charging current to 50-150 milliamps, say 10 ohms. Balancing would take just as long but no cells would be overvoltaged. If the charger cuts off at that current, put another resistor directly across the output to increase the load. The charger may never shut off then, not a problem if the current through the batteries is limited to what the BMS can handle without overheating.

Balancing can take a couple of days with as-shipped Headway cells, but if you have manually balanced them probably a few hours would suffice, once a month.
 
I have done business with them and have been pleased with their service. I purchased a 72v charger, bms, and some loose cells. The items shipped promptly, were well packaged, and the vendor (JimmieD on this forum) was helpful and courteous.
 
What to expect from 24 headway cells ? the spec's are all over the place from what i read. Tons of data sheets none match Would like to hear in service use and experiance with them? The cells are 38120 s . My pack 2p x 12 s
For the record all cells prebalanced before service.
 
You must log in or register to reply here.
Back
Top