DIY Electric Car Forums banner

Zivan NG3 Charger Algorithms

8K views 13 replies 5 participants last post by  jdeurisco 
#1 ·
I'm getting ready to send in my NG3 for conversion from AGM to LiFePo4. I just read a great thread on the modifications to be able to program it myself but I'm not ready to tackle that yet.

It is a 288V charger. I went from 24 AGM to 90 90Ah Winstons. At this point I plan on bottom balancing and going without a BMS although that could change.

The email I received from Zivan had charts with some of their charging algorithms. There are 300,500,500a and 500b series algorithms. He said I would need to pick which one I want. I'm leaning towards the 512b but could really use some advise. I didn't find these charts anywhere else. I attached the 500b series. I can post the rest if anyone wants them.
 

Attachments

#2 ·
Hi, did you get any answers. i too need to know the best algorithm for 38x 90 lifepo batteries @120v application. I have been told constant voltage constant current is the best but does this coincide with the 512b?zivan sent me the 512 or 520 can anyone enlighten me to the best algorithm to tell zivan?
thx

I'm getting ready to send in my NG3 for conversion from AGM to LiFePo4. I just read a great thread on the modifications to be able to program it myself but I'm not ready to tackle that yet.

It is a 288V charger. I went from 24 AGM to 90 90Ah Winstons. At this point I plan on bottom balancing and going without a BMS although that could change.

The email I received from Zivan had charts with some of their charging algorithms. There are 300,500,500a and 500b series algorithms. He said I would need to pick which one I want. I'm leaning towards the 512b but could really use some advise. I didn't find these charts anywhere else. I attached the 500b series. I can post the rest if anyone wants them.
 
#3 ·
The 512, 512a and 512b are all very close. I'm really not sure which would be best or if it would make any real difference. If I am reading it right, the 520 doesn't come close. Also, if I'm reading it right, the 512 series doesn't list a V1, V2 or V3 because it is in constant current during that time. I can understand there being differences if you are going to use a BMS to control the charging or not. At this point, I don't plan on doing that, so 3.65vpc during CV is about all the higher I think I want to go. Hope that all makes sense.
 

Attachments

#6 ·
I personally would select the #501b algorithm. The V1, V2 values are only used if the pack is over discharged and is there to gently bring up the pack voltage to a safe range before full power is applied. The current in this region is the Ah capacity divided by 30 so 3A in your case. Regular charge rate is Ah/1 or 90A or the charger's maximum. V3 doesn't matter so it isn't listed. It is a hold over from the lead acid algorithms. V4 is the saturation voltage. When this voltage is reached the current tapers to I3 which is Ah/80 or just over 1A for your pack. The charger will shut off after 1.5hours after V4 is reached unless your charger can shut off based on voltage and current in which case it will shut off when just over 1A is reached or 1.5h which ever comes first. If it is like my old NG1 and NG3 it will just taper the current back to just a few mA and hold it until the S3 timer times out. If you charge to 3.65v then you will be overcharging your cells. That is why I recommend the 501b algorithm. The 501 algorithm would do basically the same thing except it looks like it terminates the charge as soon as the current drops to Ah/80 which is preferred to keep from overcharging the cells.

If you want to charge to 3.65vpc then I would select the 502 algorithm. It charges to 3.65V, holds that voltage until the current drops to 3A for your pack and then terminates. If all your cells are good you won't need to balance them after an initial balance except for maybe once a year, if that. At least that is my experience with my TS pack from 2009.

Since my chargers won't terminate based on an ending current I charge to ~3.45vpc. The choices I had at the time were limited to 3.65vpc options so I chose a 19 cell setup and then installed 20 cells in my rig. I use the voltage calibration pot to make minor adjustments to the ending voltage.

A fellow Gizmo owner in FL tried a bottom balanced pack with the same charger and algorithm I have and found that there were a couple of cells which routinely went high so he changed to a top balanced pack like I have. Having the charger end when the current drops to 0.01-0.05C would be better for a bottom balanced pack. My charger is something like 0.0005C which is way too low for the normal charge procedure.

HTH,
 
#7 ·
Thanks for clearing some of that up for me David. Sorry I didn't reply earlier. I've been a bit tied up. I just got a chance to check out your blog. Very cool. I think I'm going to follow your advise and go with the 501b algorithm. I have been carefully charging them with the existing algorithm for AGM. It actually looks pretty close. It goes into final charge at about 3.60vpc. Because I haven't balanced the pack yet, some of the cells were getting up around 3.8 while others were only around 3.45 so I quit for now. Sending it in will give me time to put the cells in parallel and do a top balance. I like the idea of only doing that once a year or so. I'm not set up to swap between series and parallel very easily.
 
#8 ·
If you have power supply you can individually charge the low cells a little or put a resistor across the high cells to bring them down. There is really very little energy between 3.45V and 3.8V.

The benefit I found with reprogramming my charger is that the end of charge sequence is rather short. With the lead acid algorithm my charger would lower the current way too early so most of my charging happened at only 6A which was the equalize/absorption phase of the lead acid algorithm. Now the charger goes full throttle until the last couple of Ah in the case of my NG1 and the last 10Ah or so with my externally mounted NG3. There is bit of a voltage drop with the DC extension cord on the NG3.
 
#9 ·
I have a nice HP variable power supply that should do the trick. It is 0-10v and up to 10 amps. Any ideas on a simple "plug" to connect to the individual battery terminals? I would like to find something like a banana plug that would fit into the 8mm threaded holes. If not that, then something that would quickly and securely fit on the head of the bolts. So far, I haven't come up with anything better than alligator clips.
 
#10 ·
The single cell 10A charger I have just has alligator clips and they work just fine. The benefit to using alligator clips is you don't have to pull out any bolts to charge an individual cell. Just make sure you only have one charger going at a time in case they aren't isolated. I know the Zivan is isolated but don't know what effect it would have on an individual power supply. You don't have to disconnect the cells from the pack to charge just one cell.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top