HV Battery Headway 38120HP LiFePo4 70S pack

T1 Terry

 

HI OBJUAN,
I recently bought 2 Super Beast Modules 48 cells each pack. I am considering buying one more to make 5s2p 140cell pack on the headway cells because of what I read in this thread about the recharge rate, you burning cells, and the temperature variation/tolerance.

I started to lean towards using the stock battery and add the Lifepo4 cells to help keep it charged. but I can't seem to figure out a DC to DC converter. Was going to get expensive i think to find a DC to DC converter and a charger for 48v (two Super Beast pack) So I went back to thinking I could just swap the whole pack and found this thread.

I also bought a Meanwell LPC-100-350 and built a charger and the 2 bulb discharger to balance my stock pack but I haven't had a chance to try it. The outlet close to the parking lot of the hotel I'm at is broken.

Have you made any progress? Are you considering more cells for 2p?

I also bought some 5s bms on Aliexpress for $2 I got on the way.

 

Thanks for replying. The the 5s bms say they are 100a still that is more than what is needed I would think divided up. I'm still trying to figure out what I want to do supplement the Nimh pack or replace it. I am working about 3hrs from battery hookup and I can drive by there on the way home or back up here on the 4th of July weekend. So I need to decide if I'm going to replace the pack or not. If I do I still think I'll need to do 2p so I don't have to maintain temp requirements on charge and discharge cold and hot. If I supplement the Nimh battery I'd keep the superbeast modules intact I think and somehow covert 48v to 240vdc?

 

I like the idea with the cell links. I think it could be modified to build parallel packs to what ever capacity you wanted and link them to build the 84 module pack needed to equal the 28 module original Prius pack, 14 x 6 LTO modules.
I guess you could even use 14 of them to match the same capacity the original traction pack had, but I think I'd liquid cool them to ensure they didn't overheat working that hard. Cell voltage monitoring would be a feat of it's own, but as a piggy back battery connected one module to each two original modules would mean you could disconnect the LTO cells if a single high cell voltage occurred to allow the balancer to sort the balance problem. The onboard Prius system would look after the original module voltages.
Lots of possibilities there, with a 40CA constant charge/discharge capability gives a 240 amp capacity either way and the most I've ever seen from my Gen 2 Prius on the scan gauge.

As far as me getting a chance to build my LTO battery. hospital trips to get my double hernia sorted and recovery time, then trying to catch up on the workload that has backed up, means it's still waiting for me to actually get a start on it

T1 Terry

 

Similar problem here as far as the medical attention unless you have private insurance and can pay for the surgery and hospital time. I was passing a lot of blood when the pain events hit but that still only got me on the second level of urgency ... it took a further 6 mths and then only got the hurry up because someone else cancelled 40 hrs before admission time and we were 1600km away on a trip back to where we originally lived. A rush trip back in the motorhome and then virtually straight into the hospital, only to wait a further 8 hrs before being wheeled into theatre .... but all the doctors and nursed were great, couldn't have asked more of them and they are under such great pressure trying to catch up.

Anyway, back to the topic.
I found this video test

and the amount the Headway cells heated under load stopped that discharge test, a poor cable joint stopped the LTO cell test. The voltage under load was the other very interesting part, the LTO cells were way out in front in both the temp and voltage performance.
I also found these, or Dinah at OSN power actually suggested them as a replacement


They weld what ever tab you want on the ends. If you are interested, contact Dinah @ [email protected] .... The price and freight to the US would be a lot better than here to Australia .... but I'm still saving my bottle tops and cans to get the $$ together to buy 168 of them to build a 2P84S pack to allow for the 180 amp charging and discharging and give me a realistic 15Ah capacity min to extend the battery assist range.

T1 Terry

 

That is very interesting.

Here's what I observed in the video and what I can compare them to on spec sheets. According to Headway's specs the cells weren't overheating yet and still had quite a bit left in them although yes, the LTO's did perferom better.

I don't see why they stopped the LiFePO4's actually because they still weren't over heating and the voltage was still basically the same, at least during the portion we could see of the video.

Another thing I noticed is that the LiFePO4 cells have a higher Ah rating and higher voltage and yet they are nearly the same weight as the LTOs.

Also it doesn't seem they started these at a full charge, so there's no way of knowing at what part of the charge they were at except we can subtract a bit less than an amp hour ever minute since they were running at 50A (but not quite 60A). The LTOs obviously were not fully charged and we never say the initial voltage of the the LiFePOs nor did their loaded voltage change. It seemed to actually go up a little under load over a couple minutes.

_________6xLTOs:___4xLiFePO4s:
Start V____14.95V ____ ?V
2 min V____13.4V_____11.75V
End V_____12.9V _____11.75V
No load V__14.0V_____13.4V
End Temp__85°F/29°C__100°F/38°C

Nominal V___13.8V____12.8V
Full charge__16.8V____14.6V
Full Dischrge__9V_____10V
Max Allowable Temp
Discharging___60°C____60°C
Max Allowable Temp
Charging_____60°C(?)__45°C

The LTO's are the better battery but I don't see LiFePO4s as being terrible.

 

The part I found interesting was the voltage under roughly the same load. Volts x amps = watts, the true power being made available at any given moment. The other take away was the cell temperature rise, the LiFe cells were doing it harder than the LTO chemistry cells, this affects cycle life more than anything else.
Keep in mind, the outside case temp is a lot lower than the centre of the cell, with the original NiMh cells the KOH acts as a heat transfer fluid and moves the internal heat to the outer cases, this doesn't happen in cylindrical cells because the two plates and the separator are a continuous item wound up tightly inside that cylinder case. If the separator material exceeds its melting point, that area of the cell stops functioning so capacity is lost. That heat in the specs is the max cell heat and measured at the centre electrode, the positive terminal, this is rarely exposed for heat measurement because there is a small cell protection circuit under the cap that disconnects the cell if an over or under voltage is sensed.
What they have available in the laboratory isn't the access the end uses has available regarding cell internal temp etc.

I was disappointed they didn't fix the poor cable crimp on the LTO test and repeat it so that external heating didn't reflect the actual power produce or the cell temp.

T1 Terry

 

What would be nice is a comparison to what they would be replacing. How much do NiMH cells heat up? My guess is a lot. When I drive in hilly terrain I can tell the hybrid system starts lowering charge and discharge loads from from the battery quite quickly. Even bench charging cells they seem to bloat from off gasing quite easily even on a very low amperage charger. But without an actual test and comparison we will never know for sure how they line up. All I can find is Tenergy saying their NiMH cells can heat up to about 55°C under a "rapid charge" which seems to be in about 2 hours or at 0.5C rate (namely 500mA on a 2,000mAH cell).

Mind you the LTOs also were in 6 cells giving them more surface area to cool off too instead of just 4. The LiFePO4 batteries being sold for hybrid battery replacements are pouch style.

I'm not too sure that the center of the cells is that much hotter than the surface since there doesn't seem to be any cooling to take that heat away other than just being exposed to the air, and the temperatures were not that far from room temperatures. When the manufacturer says don't go past 60°C it wouldn't make sense to also say that measurement has to be taken from a place that's impossible to measure. The manufacturer might as well as not include any recommendation to safe operating temperatures at that point.

 

I like the last bit, all cells no matter what type can not have their internal temp measured accurately, the temp spec is the manufacturers "get out of jail free" card if a cell doesn't meet the expected cycle life.

I'm looking at full immersion cell temp control when I build my 55Ah LTO cell battery using one of the "E" fluids designed for the job so I can do away with he noisy fan and hold the cells at a more stable temp while ensuring no moisture gets anywhere near the terminals or busbars so oxidisation is also minimise. Just seems easier than trying to stop a conduction type cooling/heating system from leaking into the battery pack and a heap easier to get full heat transfer contact. Just set the fluid pressure/boiling point combination below the desired max cell temp, then any hotspots will induce a change of state and attempts to vaporise the "E" fluid, absorbing a lot more heat energy without an actual temperature rise.

The long cycle life of LTO cells compared to LFP cells and the ability to absorb high current as well as deliver high current without a significant voltage change is another real plus ..... the added volume and $$ cost if the real negative side, but hopefully the cycle life and performance will out weigh the negatives.

T1 Terry

 

The LTO cells deffinitely have certain advantages, and they deffinitely interest me. The main advantage I see is their low temperature performance. LiFePO4 and Li ion cells die if charged when frozen. Where I live temperatures can get to -40°C/-40°F in the winter. It was still snowing just a couple of weeks ago during our summer. LiFePO4 cells will need a battery heater where I live.

The problems I find is that for what I do for a 12V battery, both 5s and 6s don't seem to be optimal. In the video the 6AH battery was run at 50A for 4 minutes, or 200A minutes, more than half, and the resting voltage after that was just at 14V. But the Prius and Avalon will sometimes hit 14V but a lot of times charge to 13.5V. How is a battery that rests at 14V supposed to charge when the charging voltage is that or lower? But a 5S would be too low of a voltage because my amateur radio doesn't like to run below 12.5V., which could very well be the case with a 5S arangement.

For a hybrid battery, however, I think the LTOs would work pretty well. But again, the nice thing about the LiFePO4 packs out there is they give you a whole lot more EV range since the charge characteristics make them use about 80% of their capacity (10% to 90%) instead of 50% like with the NiMH cells. From what I've seen in videos the LiFePO4s also outperform the NiMHs already.

 

The catch with the LFP cells is the recharge current on regen and light to medium braking. LFP is really limited to 0.5 CA charging, 1 CA will shorten the cycle life and 5 CA will most definitely shorted the cycle life. Even at 5 CA that is only 30 amps into a 6Ah cell, the Gen 2 Prius will regen at over 120 amps. The LTO cells will handle 10CA, so 2P 6Ah LTO cells would handle a 120 amp charge as well as as 120 amp discharge much better than the LFP cell.
The Yinlong 9Ah that is probably more like an 8Ah when being push hard, configured in a 2P arrangement could handle 160 amps charging or discharging and work as a direct replacement for the original NiMh modules and still return a great cycle life.

As far as the 12v start battery, I use a lot of tired 90Ah and even some tired 60Ah LFP prismatic cells configured 1P4S and they work brilliantly as a start battery as well as a portable power pack for the 12v chainsaw. The 1P4S pack drops straight into where the Gen 2 lead acid battery was, and no more dead start battery problems .... although when it drops to single digit figures C* around here we all think we are gunna die but on the other end of the scale, we do suffer days at 48*C plus during summer (I think that's around 120*F)

T1 Terry

 

I would die at 48°C. I think I'm going to die when it's at 40°C (104°F).

Ya, I think the current hurts the life in the LiFePO4's. But from what others have said, the LiFePO4s normally have double the longevity of NiMHs. But using them as a hybrid battery brings them to about the same longevity, the difference being they give about 3 times the capacity for about half the weight and at about the same price.

 

When it gets up around the 55*C on a houseboat roof I reckon I get real close to toppling over and have to get down while I still can :lol: Had one job where the roof was so hot at 10 am, I could touch it or kneel on it without my protective closed cell kneeling board. I put down a piece of heat shrink and when I picked it up again it had shrunk so small it wouldn't fit over the wire. Tried to pick up the tools and they were too hot to touch, so I left it all up there till that evening :lol: You certainly drink a lot of water when it gets hot, the workshop mezzanine level is too hot to walk around up there in summer, the dizzy spells hit after a minute or so, even with the big roller door and the roof vents open, even at 6pm the western wall is too hot to stand near, so we don't :lol:

The LFP cells increase their internal resistance significantly after being fast charged too often and reach a point where no current flow in occurs under 4vdc, yet as soon as the charging is removed, the cell voltage drops to 3.2vdc or less. It is the electrolyte heating that causes this problem, so if there was some way to keep the cells cooler than 30*C they probably would probably handle the job as a NiMh replacement and give a long service life.
The NiMh don't do that bad as far as service life goes, if the KOH electrolyte is refilled in each module, they come back to life. The pressure relief valves cause the most trouble, the pressure gets so high the seals on the terminals fail and vent the KOH leaving that crud building and attacking the module links.
Making a recycling KOH vent/recovery system with a cooler so only cooled KOH returned to each module would really extend the service life ..... but I can't see that kit hitting the market any time soon :lol:

T1 Terry

 

I've been considering the Headways cells for a Prius C 2014. I wanted data so I wrote a script to gather the data from my routine drive to and from work. The script still has some errors, but I have it shared here if anyone is interested. for the prius you will need to change a few things. github "SeniorPrius". I can't link yet since I just created my account.

After gathering the data, I processed it to get a distribution of RPMs, voltages, current, power etc per the OBDII data:

As you can see from the picture the voltage sort of stays at 1.35V per cell.

Using the same script, I can play around with the number of cells and for example if for the 10 modules on the Prius C, I used 50 cells per your arrangement, I would get a voltage distribution that looks like this:

the voltage would land at around 3.2V which is like 5% SOC on LiFePO4. I was thinking if I reduced the number of cells to 47, I would get a better number, around 3.4V.

What do you guys think?
As someone else had mentioned, that would mean placing the 10 voltage taps on the Prius C right at 4 cells, and then having an extra 7 cells after the last measured cell. I would then install small active inductive balancers (overlapped) in between say every 10 cells or so to keep the solution cheap. Amazon sells a 12S for example:

I'll post images of the distributions tomorrow when I can. I gotta post 4 things and wait a day before I can post a link apparently.

 

I was hoping to figure out what happened to OBJUAN's use of the headways. Any update on that? did you just go to a different battery?

 

Battery hookup has this other battery: A123 3.2v 31ah LifePo4 cells. It sort of almost fits but it seems to be a great electrical fit, just really darn expensive.