The LiFePO4 Headway cell 38120P TEST REPORT inside

[ZapPat]

Anyways, I only used 0.26Ah but here is what the CA repots:

Max Amps: 33.93 A
Min Volts: 66.6 V

Assuming max amps and min volts occurred at the same time, I hit a peak of 2.26kW. Probably the case, as I did see ~2.15kW up a steep hill.

Resting pack voltage 5min afterwards: 79.9 V (starting from 87.9 V hot off the charger)

Assuming the numbers are right, 66.6v / 24 cells = 2.775V/C at ~3.4C discharge rate.

Maybe you should do a load test using two different loads, because so far these cells may look like they have the characteristics of the old cells, meaning over 11-13 milli-ohms per cell. The only good results we've seen are from "biggs" forum member, and he has only posted 5 times in all, and not since the last page of this thread. Are you still here, biggs, and how are your cells doing in real life?

However, there are quite a few variables that could seriously through off the results of my quick calculations: Beginning of discharge (surface charge), use of only one voltage/current point, unsure if current and voltage figures were took at the same operation point...

I used your 79.9V float voltage (at 0 amps) along with the 66.6V at 34A to get an idea of the internal resistance. I reduced the float voltage down to 78V just to reduce some of the unavoidable error produced by using a no-load (0A) point. So...
R = (78V - 66.6V) / 34A
R =~ 335mOhms (for the pack)
Rcell =~ 335/24 =~ 14mohms per cell.... ugg!

Of course the BMS introduces some resistance itself, but it should not be too high since it is supposedly made to support high current discharges.

Now if we could see a good two-point operation test of these new headway batts, it could either confirm this mediocre result or (we hope) bash it into the ground.

 

[voicecoils]

Now if we could see a good two-point operation test of these new headway batts, it could either confirm this mediocre result or (we hope) bash it into the ground.

Cool. I haven't read the whole thread, but now I get the whole internal resistance thing.

I have a feeling my wiring/connectors may be adding some resistance to. It's mostly 10ga and soldered on 45a power poles. I've ordered a proper crimper and new contact pins and will re-do them. At least 1.5m of battery wiring.

When it's daytime I'll try to do a longer ride and report back on results. I've just put a nice molex style connector on my hall wires so they wont be coming loose now.

I'll try to ascertain voltage drop at 10a, 20a, and 30a. Hopefully I can repeat it at something like 2/4/6/8ah consumed. That should provide some further data points to ponder.

 

[Kurt]

I am not fully up to scratch all the the tec talk but if after the battery break-in period as in a couple of charge discharges and and some balancing if my pack doesn't perform I wont be a happy camper. I would expect mine to at least perform as well as a ping pack if not better for the money.

What kind of test do you want me to do using CA I can pull full throttle a moderate hill and load a 36v 20ah pack to 48amp's. Do you just want the lowest voltage at this point And the voltage at rest or do you need the voltage drop at two load points say 20amps and 48amps?

I did get an email to say the packs wont perform to full potential until a couple of charges/discharge/balancing cycles.My guess is that would have more to do with there ah capacity .

I wont pass any judgement until I test them for my self.

All my wiring is very capable and a little over rated and over done.It tends to be how I do things. Short cable runs. Fully crimped and soldered connections and where I can't run the gage cable I want due to size I run two runs of thinner cable .So there shouldn't be any issues with my data on CA and nothing I can blame my bike for.

I don't expect the World from them but I expect they should perform as well as other battery's for the same price.
Kurt

 

[Kurt]

I got my headway battery today 36v/20ah made up of 24 cells.I also got my 5304 and controller When I opened the cardboard box from headway I found a long pack two rows of 12cells on top of each other.The battery's were wrapped in a very cheap and tacky stiff paper wrap with clear tape holding that on. Very cheap and nasty.The wires coming from the pack were at least nice and thick. I threw the pack in the charger and while handling the pack I noticed if I pressed on the wrapping the charger would go out

Time to cut the cheep paper off and see what's going on underneath.The cells are HW 38120 10ah. they have the brown wrapping.I found the problem in about 30 sec one of the tabs with about 20 spot welds on it had come away from the battery. Very pore spot welding.After talking with my supplier I did a very fast soldering job with a large iron.Its nice and strong at that point now But I don't really trust the other spot welds.

At this stage until I get my 5304 back from the bike shop who is truing the wheel. I would have done it my slef but it had a small flat spot in it .You should have seen his face when he grabbed the wheel I dont think he was expecting the 12kg .

So not a good start on the headways like most things if you want it made correctly usually you have to do it yourself.Because my suppler will warranty it unless I do something silly My plan is to get the pack up to scratch.

I loaded the pack up to 25amps and the Voltage went down to 38v from about 40v on CA.The BMS looks well made and very solid big heat sink.

I was thinking about some of the options for improving the cell connection what about braded wire like the rc guys use? I just don't trust the spot welds.
I also want to do something to stiffen the pack up a little there is nothing bit a tiny strip of hot glue keeping the cells joined to each other. I was thinking any flexing would put load on the spot welds.I was thinking even just some sheets of aluminium on the top and bottom. I don't want to box it in to much if there is any heat that could be a problem.

I should have the bike going wit 5304 buy the weekend so I will load them up and report back with the results.

 

[voicecoils]

Kurt, Good to see you got your pack! Looks like the older cells from the brown paper. The support from John is key with these packs IMHO. We have few options in this price range so having a supportive dealer is really good. I can't wait to get my ebike back on the road so I can get a full years worth of use out of them under warranty. After that...

Do you have a link to your 5034 build? I'm curious about it, since I just got the same motor myself

Also, do you have a link to the flickr pics of the pack? I'd like to see some closeups/big pics of the BMS and tab welds.

 

[biggs]

Anyways, I only used 0.26Ah but here is what the CA repots:

Max Amps: 33.93 A
Min Volts: 66.6 V

Assuming max amps and min volts occurred at the same time, I hit a peak of 2.26kW. Probably the case, as I did see ~2.15kW up a steep hill.

Resting pack voltage 5min afterwards: 79.9 V (starting from 87.9 V hot off the charger)

Assuming the numbers are right, 66.6v / 24 cells = 2.775V/C at ~3.4C discharge rate.

Maybe you should do a load test using two different loads, because so far these cells may look like they have the characteristics of the old cells, meaning over 11-13 milli-ohms per cell. The only good results we've seen are from "biggs" forum member, and he has only posted 5 times in all, and not since the last page of this thread. Are you still here, biggs, and how are your cells doing in real life?

However, there are quite a few variables that could seriously through off the results of my quick calculations: Beginning of discharge (surface charge), use of only one voltage/current point, unsure if current and voltage figures were took at the same operation point...

I used your 79.9V float voltage (at 0 amps) along with the 66.6V at 34A to get an idea of the internal resistance. I reduced the float voltage down to 78V just to reduce some of the unavoidable error produced by using a no-load (0A) point. So...
R = (78V - 66.6V) / 34A
R =~ 335mOhms (for the pack)
Rcell =~ 335/24 =~ 14mohms per cell.... ugg!

Of course the BMS introduces some resistance itself, but it should not be too high since it is supposedly made to support high current discharges.

Now if we could see a good two-point operation test of these new headway batts, it could either confirm this mediocre result or (we hope) bash it into the ground.

hello i am still here,

at this time i am working on the BMS and i will finnish it this week. the only thing that i can say is that at my first test with my pack (24s4p) the performance of the bike is not bad. we draw peak current of 350A and 80A continue. Atn the continue current the voltage goes down to 70V. And as i have measured the resistance of the cell i have an overall resistance of 120mOhm. the peak power of this pack is about 21KW. that was with the old cells. With the new one, i have got it from victoria, i changed only 36 cells in my pack, i have 93mOhm resistance and the peak power of 27KW. If i change all cells to the new one i will have 48mOhm and 54KW peak power. ( Peak power = U*U/(2Ri) ). for longtime test i have to build or order an cycle analyst. the furture will bring more results. my new cells are with screw tabs and they are sealed with a blue cover.

so long

biggs

 

[voicecoils]

at this time i am working on the BMS and i will finnish it this week. the only thing that i can say is that at my first test with my pack (24s4p) the performance of the bike is not bad. we draw peak current of 350A and 80A continue. Atn the continue current the voltage goes down to 70V. And as i have measured the resistance of the cell i have an overall resistance of 120mOhm. the peak power of this pack is about 21KW. that was with the old cells. With the new one, i have got it from victoria, i changed only 36 cells in my pack, i have 93mOhm resistance and the peak power of 27KW. If i change all cells to the new one i will have 48mOhm and 54KW peak power. ( Peak power = U*U/(2Ri) ). for longtime test i have to build or order an cycle analyst. the furture will bring more results. my new cells are with screw tabs and they are sealed with a blue cover.

Nearly 90A per string peak! wow. What is the voltage drop at this level, ~60v?

If that were the case, I would theoretically be able to get 5.25 kW out of my pack , though I don't think the headway supplied BMS would like this.

Keep us updated with your results. Can you photograph your new cells and the current pack?

Where are you located Biggs? Perhaps I could lend you my stand alone CA for a while, if you are able to use your own shunt (as the one included is 45a continuous, 100a peak)

 

[biggs]

at this time i am working on the BMS and i will finnish it this week. the only thing that i can say is that at my first test with my pack (24s4p) the performance of the bike is not bad. we draw peak current of 350A and 80A continue. Atn the continue current the voltage goes down to 70V. And as i have measured the resistance of the cell i have an overall resistance of 120mOhm. the peak power of this pack is about 21KW. that was with the old cells. With the new one, i have got it from victoria, i changed only 36 cells in my pack, i have 93mOhm resistance and the peak power of 27KW. If i change all cells to the new one i will have 48mOhm and 54KW peak power. ( Peak power = U*U/(2Ri) ). for longtime test i have to build or order an cycle analyst. the furture will bring more results. my new cells are with screw tabs and they are sealed with a blue cover.

Nearly 90A per string peak! wow. What is the voltage drop at this level, ~60v?

If that were the case, I would theoretically be able to get 5.25 kW out of my pack , though I don't think the headway supplied BMS would like this.

Keep us updated with your results. Can you photograph your new cells and the current pack?

Where are you located Biggs? Perhaps I could lend you my stand alone CA for a while, if you are able to use your own shunt (as the one included is 45a continuous, 100a peak)

I am located in vienna Austria. I think i would build my own CA because i want to have a grafic display and i don't want to have an CA and for the BMS with the new ltc6802 i need an extra bordcomputer for configuration of this chip. my analog BMS is ready know and i will make the pcb board. with this bms i can shunt 12 A each channel.

i will keep you informed.

bg
biggs

 

[voicecoils]

I am located in vienna Austria. I think i would build my own CA because i want to have a grafic display and i don't want to have an CA and for the BMS with the new ltc6802 i need an extra bordcomputer for configuration of this chip. my analog BMS is ready know and i will make the pcb board. with this bms i can shunt 12 A each channel.

i will keep you informed.

bg
biggs

no worries, the offer is there if you run out of luck

100k post?

 

[Kurt]

Just a small update on my headway pack 36v/20ah.

I have been using this pack for a couple of weeks now its powering A 5304 with 48A controller no current limiting.Its ridden over a good mix of hills and flat sections with little to no peddling and often gets blasts of 1600w from a a dead start or up very steep hills.The bike is a heavy izip/ezip mongoose style steel bike with baby seat and 11kg child + my 75kg back and x5 so about 130kg or 260lb all up.

The only modification I have done to the pack is to reconfigure the cell arrangement to make the back a more practical shape.Today I ran the pack down to the low voltage cut-out on the bms.The good news was the LVC on the bms cut in befor the controller lvc so I know it works. CA reading was just over 19ah.I was taking it a little easier on the throttle during the last ah of the pack but being a x5 it can still suck some decant wattage even when taking it easy.

I have about 7 cycles on the pack now so it should be broken in by now .I feel that about 16ah is a better self imposed cut-off point for longer battery life. Most of my riding is 30km or under so 10-12ah has that covered.The cells did about 5 mins of balancing when I first got them and haven't seen the charger balance after this.The voltage drop under load may not be as good as some of the top end cells but after the 19ah ride on x5 I was never disappointed with the performance and didn't require any peddling.

I Just need to l keep filling in the log I am keeping on the pack and give it some time as in a year or so to see if I am 100% happy.
Kurt

 

[voicecoils]

Just a small update on my headway pack 36v/20ah.

The only modification I have done to the pack is to reconfigure the cell arrangement to make the back a more practical shape.Today I ran the pack down to the low voltage cut-out on the bms.The good news was the LVC on the bms cut in befor the controller lvc so I know it works. CA reading was just over 19ah.I was taking it a little easier on the throttle during the last ah of the pack but being a x5 it can still suck some decant wattage even when taking it easy.

Very good to hear! 19ah before cut out from a 20ah pack sounds great. Do you know the before and after pack voltages, at 0Ah and 19Ah? Did you feel a drop in power by then end?

I still plan to leave the packs on the charger overnight at least once a week to give them time to balance.

 

[Kurt]

At about the 18 ah point you could feel it didn't have the same kick as in the beginning.You only really notice when trying to go full throttle up a hill instead of getting a blast of 1700w or 1800w you get more like 1500w just because the voltage under load starts to drop off in the last 2ah.You will notice it but you have to consider I am running a 48A controller.If you ran current limiting or a 20ah controller you could get close to if not the full 20ah from the pack. Mine only hit the LVC when I gave it full throttle up a hill at the 19ah mark.Perhaps i could have kept riding it keeping the wattage under say 800w but I didn't see the point.I just limed home 1/2 throttle for 500m.

42v fresh off charger at 0ah at 19ah after a couple of min rest it was at 37.4 this voltage drops fast under load when the battery is close to flat.I think the LVC is around 30v cuts the pack off you need to unplug the pack then plug it back in to reset it.

One thing I have been looking for is any kind of cell heating on both the bms and the cells them self I cant feel any more heat than the ambient temp.Perhaps I am not pushing them hard enough with a x5.When people lab test the cells at say 30A or 40A constant perhaps they will get hot but on a ebike your not going to jump on and pull a full 40A for 30min constant unless your entire ride is just one big hill.


John is getting back to me with the cell specs from headway as I want to check on a couple of things. One is the chargers voltage is only 42.5v seemed low at first but I have a feeling that the nominal voltage of the headway cells is a true 3.2v and a little lower than most lifpo4 and don't need to charge them as high.the 2nd thing I wanted to know there live cycle rating to.
kurt.

 

[voicecoils]

42v fresh off charger at 0ah at 19ah after a couple of min rest it was at 37.4 this voltage drops fast under load when the battery is close to flat.I think the LVC is around 30v cuts the pack off you need to unplug the pack then plug it back in to reset it.

kurt.

My chargers charge the packs up to 44.0v, but after just a few min off the charger the packs are down to ~43.8 and keep dropping bit by bit over a few days. I'm not sure what it would level out to after say 1 week.

Anyways, for your application at least it sounds like the pack is perfect. If your controller only goes up to 48A and the batteries will happily give supply it for a full 19ah then that seems great. Do you know how many Wh was pulled from the pack? A single 3.2v10ah cell should be able to provide 32Wh, so your cycle analyst would have shown ~690Wh or a bit more I suspect. Shame the BMS cut out on a hill :wink: but half throttle is better then nothing!

We have different charges (from the pic you showed) but be very careful about polarity when you plug them in. I just blew one of my two chargers by being careless with my the way my powerpoles were arranged. Having red/black & blue/green powerpoles is a bit too confusing! My charger doesn't have any reverse polarity protection it seems. I'll make a thread about it maybe someone clever has a easy fix. Here's a big pic of my charger: http://www.concordski.athena.thinkhost.com/lifebatt/DSC_0517.JPG Looks very simple from the top but there are a bunch of surface mounted chips and other components on the backside.

Cheers!

 

[Ypedal]

A trick with andersons.. add an extra housing (without any wires at the end ) at 1/4 turn to prevent plugging in the wrong way.. glue them in place.

 

[Kurt]

You could also try deans plugs they are imposable to plug in the wrong way.I have swapped to them for everything .They have a nice tight fit good current rating and compact.

 

[voicecoils]

Those are both good tips. Thanks guys. I should stop steering things OT. I'm firmly invested in 45A powerpoles now so I will stick with them, and if needed with the X5 will double up with 8-10 gauge on the battery side. Ensuring I don't plug things in wrong though IS key.

Is there a standard positioning of the plugs? If you're holding a wire that terminates in a plug in front of you, should the positive (red) be on your left or right? Maybe there is no standard way.

I'm leaning off topic again, damn!