Headway 24s1p - Best way to do it?

[El_Steak]

I want to go with an 88v / 50amps setup for my bike. 10ah is enough, but I need to be able to carry the charger(s) on the bike. Headway cells are very tempting but I'm not sure which way to go with all the packs, BMS and charger possibilities. I came up with the 3 ideas listed below, can I get some comments/suggestions from the experts?

Idea 1:
Get a 48v 10ah pack (like the one YPedal got) and a 48 volts LifePo4 charger. Buy an additional 8 loose cells and 8 voltphreak single cell chargers.

Pros:

• Can go anywhere between 58 and 88 volts depending on how many loose cells I use.
• With single-cell chargers, balancing becomes a non-issue on the loose cells
• Fairly easy to set up, no BMS to buy and wire
• If a cell goes bad, I can just discard/replace it

Cons:

• No discharge protection on the 8 cells, I would have to be very careful and use a conservative cut off voltage on my Cycle Analyst
• 1 x 48v charger plus 8 x 3.65v is a lot of charger to carry around (I have to recharge at work)
• Many things that can go wrong with the 8 small chargers

Idea 2:
Build my own pack from scratch using the red plastic blocks, buss bars and the Goodrum / Fechter BMS

Pros:

• Its fun
• I can configure the pack any way I want

Cons:

• A lot of work
• Many opportunities to screw up the BMS, cells, etc.
• I haven't found a cheap portable charger/power supply (2 amps is enough) that I can use to charge at 90v
• Maybe I can wire 2 36v chargers in series?

Idea 3:
Buy 2 x 36v10ah packs from evcomponents (the ones in a plastic or aluminum case) and 2 x 36v chargers.

Pros:

• No assembly required other than a series adapter
• Nice case
• 36 v chargers are cheap and easy to find
• BMS included

Cons:

• Packs are limited to 20amps discharge so I'd need to bypass the BMS for discharge and only use it for charge
• Again, need to monitor the pack closely and use a conservative cut off voltage on my Cycle Analyst
• The batteries in those packs are soldered not screwed together. Replacing a bad cell could be more trouble.
• I've never seen the inside of those packs so I don't know how thick the wires are, how good the connections are, etc.

 

[dnmun]

the tabbed headway cells in that pack are not soldered together, they are spot welded in series through the tab. you can replace the bad cell by cutting it out and leaving the adjacent tabs long and then solder the tabs of the new cell to the remaining tabs when you install it. not too hard. they should last longer than you though so i would not worry about replacing one.

do you know for certain that the 36V BMS limits at 20A? that could be cured by reducing the shunt resistance in the BMS.

jeremy harris and nick(tiberius) have two threads on using the meanwell power supplies as CCCV chargers. you could live simple and just buy one of the 48V power supplies, and then if you had 2 of the 36V packs, you could charge them in parallel. you can adjust the voltage output from 48V down to 44V. then charge both bateries at the same time with the one power supply and get 5-6A charging on each.

as i understand the argument, you should install a diode in parallel with each battery in series so that if one of the batteries shuts down the output FETs on the other BMS would be exposed to the full voltage of the two packs. maybe someone else who knows electronics can illuminate that argument.

i had another idea about using 2 of the 36V packs in series, bypassing the output FETs in the BMS by taking the main power directly off the battery terminals before the BMS.

in this case i imagined that the lower pack would supply the 36V for the controller so that the controller would need to have the main power and S/D caps replaced with the 100V1000uF caps like i did on the 72V upgrade to the 36V 9FETs controller. this is because the controller FETS would see the full 88V.

the power from both packs in series would run directly to the controller main power busses, and the controller circuit current would come off the lower of the two packs. if the lower pack shuts down because of LVC them that would shut down power to the controller and protect the lower pack.

the problem is with the upper pack needing to signal LVC if one of the cells in the upper pack go low.

in that case a small mechanical or solid state relay which was driven by the output from the BMS of the upper pack could be used to short the ebrake line to ground if the upper pack BMS detected LVC and shut down the output. so that would be a normally closed relay which would be held open by the voltage output from the BMS of the upper pack. if the upper pack shuts down, the relay closes, and that shuts down the controller, protecting the upper pack.

to monitor main pack current through the controller FETs you could use the CA to throttle back if the current demand was too high, or simply ignore it and push the current anyway.

i also am thinking of how to build a DC converter to run off the lower pack and supply the 15V for the controller current.

in that case the lower pack BMS would not be able to shut off the controller current if it detected LVC, so you would need a second relay. or maybe an optoisolator with a big power transistor inside to switch 2 of the ebrake lines if either of them detected LVC.

so you would be able to run the two 36V packs in series without running the main power through the output FETs in the BMS so the BMS would not get hot.

also both the packs would still have LVC protection and the output FETs in either BMS would not be subjected to the full 2 pack voltage if one of them shut down so no need for diodes in parallel. all just ideas.

 

[El_Steak]

Thanks for taking the time to reply with so many details and ideas.

I must admit that when it gets to diodes, relays and optoisolator, a lot of this gets beyond my current level of knowledge and skill.

The more I think of it, the more I would prefer to go with a simple setup with one pack, one BMS and one charger. The pack building and BMS (fechter/goodrum) are no problems, but the charger is hard to find. I had hopes of being able to wire 2 cheap 36v SLA chargers in series but I read in another thread that there are some risks with that. I will leave this unattended to charge during at night time in my garage and I don't want to burn the house down. Especially since my bedroom is right above the garage

Most of the 90v capable chargers and power supplies I found are big, bulky and heavy. I don't want to be carrying this around on my bike and I don't really need 10amps charging rate. 2 is plenty.

I did find this charger: Soneil 7203SR that looks interesting. Its only 1.2amps but thats enough for me to fully recharge at night and during the day at work. They do mention that it is "equivalent to a 4Amp heavy transformer charger in charging time" but I'm not too sure about what they mean. I looked up the specs and it maxes out at 87.6v. A tad short for a 24s setup, but there might be a pot inside that would allow me to bump it up. If not, 23s is fine with me.

I also like that fact that the charger is small (Length - 6.9" (174 mm), Width - 3.4" ( 86 mm), Height - 2.3" ( 57 mm)) and super light weight (1.1 pounds). It also draws no current from the battery when connected to it but AC not plugged in. This could allow me to leave permanently connected to the pack inside the battery bag.

It's around 100$ and I think Soneil has a decent reputation for charger quality.

Do you think this would be a good candidate to use with the headway pack and fechter/goodrum BMS?

 

[Tiberius]

Hi,

I do actually run Headway 24S1P on one of my bikes.

I have them arranged as 3 packs of 8S1P (nom 24 V). Each pack has a cell level LVC with optocoupler output.

For running the bike they are connected in series. The LVC outputs pull the throttle low.
They are charged in parallel using the modified PSUs and 3 diodes. If I'm really in a hurry I can parallel up PSUs to get 30 A and do a 1 C charge
Every so often I unplug the LVC board and connect a balancer to each pack and balance the cells.

Battery pics at the bottom of this page http://www.re-voltage.eu/redbike.html
LVC board in the middle of this one http://www.re-voltage.eu/electronics.html

A 24 V pack will need about 29 to 30 V from the charger, depending on what you do about the BMS. If you go for 2 x 36 V packs, they will want 44 or 45 V, which fits nicely with a 48 V PSU. But to charge the whole 24 cells in series you need 90 V, which is getting difficult.

Nick

 

[fechter]

I did find this charger: Soneil 7203SR that looks interesting. Its only 1.2amps but thats enough for me to fully recharge at night and during the day at work. They do mention that it is "equivalent to a 4Amp heavy transformer charger in charging time" but I'm not too sure about what they mean. I looked up the specs and it maxes out at 87.6v. A tad short for a 24s setup, but there might be a pot inside that would allow me to bump it up. If not, 23s is fine with me.

....
Do you think this would be a good candidate to use with the headway pack and fechter/goodrum BMS?

It should be fine as long as you can get the voltage up. Using 23s should work too. Most Soniels have a voltage adjustment of some sort. I have some instructions around here somewhere, but for a different model. You want to crank the voltage up enough to fully turn on all the shunts. This should keep it from dropping into the float mode at the end of charge.

 

[El_Steak]

It should be fine as long as you can get the voltage up. Using 23s should work too. Most Soniels have a voltage adjustment of some sort. I have some instructions around here somewhere, but for a different model. You want to crank the voltage up enough to fully turn on all the shunts. This should keep it from dropping into the float mode at the end of charge.

I think I'll buy one and try it out. Hopefully there is a pot inside if not, I'll go 23s or maybe dnmum will be able to advise on a resistor change that would do the bump. He really seems to know is way around chargers.

I went through the specs and this charger seems to do a bit more than the standard ones. Especially stage 4 were it drops down to a lower voltage (81.8V) do you think this would cause any trouble with the BMS ?

Stage 1: Deep Discharge Charging Pulse Mode: LED Flash
The charger starts charging at 0.5V and give pulse current up to 5V. This has effect of removing loose sulphation formed during deep discharge state of the battery.

Stage 2: Constant Current Mode (CC): LED Orange
The charger changes to constant current 1.2A. When the battery voltage reaches up to 87.6V, the charging stage changes from CC (Constant Current) to CV (Constant Voltage) mode.

Stage 3: Constant Voltage Mode (CV): LED Orange
In this stage the voltage of each cell in the battery is equalized. The charger holds the battery at 87.6V and the current slowly reduces. When the current reaches at 0.5CC, this point called the Switching Point. The Switching Point is one of the great feature of this battery charger that it can adjust the current automatically according to battery capacity. Other chargers are not capable to adjust the current automatically.

Stage 4: Standby Constant Voltage Mode: LED Green
The charger maintains the battery voltage at 81.8V and current slowly reduces


Thanks.