Page 13 of 108
Posted: Tue, 03 Mar 2015, 05:06
The graph above shows the storage life of Lifepo4 cells at elevated temperatures without even using/abusing them
Thanks lopezjm2001. Can you link the full article that came from?
Posted: Tue, 03 Mar 2015, 05:31
Thanks lopezjm2001. Can you link the full article that came from?
It comes from a datasheet. I am not at liberty to say which specific Lifepo4 cell. Just to say that one Lifepo4 cell is the same as the next.
Posted: Tue, 03 Mar 2015, 13:32
Thanks for the graph.
It looks like they have 5 years worth of real data and trend lines from that. Im not to disappointed with the results. Given the temperature would need to be constant to achieve the results.
Roughly 17% capacity loss with a 25c constant. Just guessing but I don't think we would be much off that temp in QLD over the year.
Most capacity tests (particularly at very low C rates) have resulted in more than the rated capacity from the calb lifepo4 cells. So even if you were down 10%...20% after 15 years I could live with a 360 - 320AH bank offgrid
Though just thinking If we lost say 17% capacity from environmental conditions over 15 years then we would have to add capacity loss from cycling the cells on top of that.
Posted: Tue, 03 Mar 2015, 13:49
Great graph. I have a feeling that this discussion should be linked to, or continued, over in "Batteries, charging, management and monitoring".
In a years time I'll never find it here.
Posted: Tue, 03 Mar 2015, 21:43
oops, It's actually 2 years of data (days squared)
I wonder if anyone will actually do a long term test. Will short tests and trend line projections always paint the full picture over long periods.
Posted: Tue, 03 Mar 2015, 22:11
I don't understand why anyone would take much notice of this graph when we have no idea of its source, which clearly isn't a peer-reviewed journal. And nor do we have a description of the experimental protocol or any explanation of what it means by "recovered" or why it should be accurate after extrapolating by a factor of 8 in time, or why it should be approximately linear in the square root of time.
Although I can
say I've had some square roots of a day myself, when nothing goes right.
I'm guessing log of time might fit straight lines well-enough to the actual data too, and would extrapolate quite differently.
Posted: Tue, 03 Mar 2015, 23:31
weber wrote: I don't understand why anyone would take much notice of this graph when we have no idea of its source, which clearly isn't a peer-reviewed journal. And nor do we have a description of the experimental protocol or any explanation of what it means by "recovered" or why it should be accurate after extrapolating by a factor of 8 in time, or why it should be approximately linear in the square root of time.
I could not find any mention of a experimental protocol in the datasheet nor a explanation of recovered.
I would have to assume that on the seven occasions of measuring the capacity loss over two years they must have cycled the cell at least once on each occasion, hence the term recovered.
Posted: Tue, 17 Mar 2015, 16:01
All this talk about keeping batteries cool got me thinking more about the freezer idea. I was doing some searching on Ebay this morning. I was looking for a chest freezer with the correct dimensions to house my 16x400ah calb cells with.
Plan was to remove the thermostat and install a little digital unit I have laying around to covert it into a actively cooled box pick a mild temperature setting. Would require some cable access, perhaps a small wecam inside to keep tabs on the LED status of the cell top module status.
Given the consumption of the 325lt unit I have my eye on is only 500whrs /24hrs as a freezer Im sure it would be only a fraction of that at more modest temperature.
This is the one I was looking at.
http://www.ebay.com.au/itm/161639467224 ... EBIDX%3AIT
Do you think it's worth the effort. Can anyone see any issues using this freezer.?
Posted: Tue, 17 Mar 2015, 17:00
I wonder how much weight the freezer can hold?
Posted: Tue, 17 Mar 2015, 18:16
offgridQLD wrote: All this talk about keeping batteries cool got me thinking more about the freezer idea.
Do you think it's worth the effort. Can anyone see any issues using this freezer.?
Yes! Definitely worth the effort (easy for me to say since I'm not spending any time or money on it). I think this would be a great experiment. Of interest world-wide.
Your temperature probe for controlling the compressor should probably measure the temperature of the air in the freezer, or an outer part of a cell case, not the temperature of cell terminals or in the middle of the block of cells. We wouldn't want it running continuously and briefly-freezing the bottom of the cells and causing lithium plating on charging, because it is taking too long to have any effect in the middle of the cells. The probe shouldn't touch the sides of the compartment either or it may cause the motor to try to turn on and off too often.
One worry is water droplets condensing on your terminals and CMUs (and promoting corrosion) every time after you have opened the lid and let warm humid air in. Being a chest type would certainly reduce that, compared to an upright.
I was recently looking into miniature air conditioners for 19" server racks, for the purpose of LiFePO4 cooling on someone else's project. Their prices are ridiculous.
One drawback of a chest type is that the solvents used for the electrolytes of Lithium ion cells, form explosive mixtures with air over a wide range of concentrations. This is a bit like the hydrogen from lead-acid batteries. Except these mixtures are heavier
than air, and so will sit in the chest and could be ignited by a spark during maintenance, unless there is some way for them to drain out.
Now I know these solvent vapours aren't supposed to vent unless the cell gets to about 130°C, but when I used to keep my 160 Ah Sky Energy cells in a large chest toolbox, I would smell the fruity nail-polish-remover smell of these solvents every time I opened the lid. Maybe the human nose is very sensitive to these smells, due to having evolved to detect ripe fruits, and so perhaps this never represented anything like an explosive concentration. It would be good to know for sure.
Posted: Tue, 17 Mar 2015, 18:54
All very good points.
The condensation is something that did come to mind and could be a issue. I don't like corrosion and don't want to encourage it.
Being so small and totally sealed box could be a concern if there is venting.
The weight was considered but big chest freezers get loaded up with a lot of heavy items so I think it would be ok with the 220kg or so.
I would definitely only probe the air temps. I would want double redundancy on a device capable of turning the cells into a ice block. Perhaps a 2nd thermo controlled switch controlling the main power input as a backup.
I guess the good thing about traditional style air conditioner is it removes condensation and dumps it out of the room and cooles all of the components in the power room. A chest freezer will cool the batteries but dump hot air into the power room that the other components are trying to cool down in.
Split systems having built in timers so you could set it to run from say 9am - 6pm .
2kw output is typically the smallest easily available system.
I'm finding the heat created by the charge controllers and the inverter in the same room affects the overall room temp.
Just something about having a split system running that I'm not sitting under or benefiting from in any way seems odd.
short of locating the power electronics and batteries in a common room that's already going to be air conditioned for personal use like a section of the shed/garage (probably the smartest idea) There isn't much of a option. Perhaps that's just something I need to get over.
Posted: Tue, 17 Mar 2015, 20:01
Kurt, if it's just Cell temps you are controlling then if they were in a fairly insulated box with an intake and an exhaust vent, then feed the intake with one of those little personal aircons ($400-500 on eBay). The little aircon could vent outside via it's 100mm tube or what ever they use. An alarm etc. if the box got too hot due to failure somewhere.
That way you don't contribute much to the room getting hotter.
A split system seems a bit overkill for just the cells.
Posted: Tue, 17 Mar 2015, 20:37
Yes I do have a potable Air-conditioner that I picked up at a garage sale for $70. Electrolux 2.5kw system. I modified it to have two tubes exhaust and intake so it functions correctly. Link to small video I made on how to do this https://www.youtube.com/watch?v=iGTzQ2MiBUw
Though it's disgustingly inefficient and not just because it isn't a variable speed compressor.
What is often overlooked is how important the size of the condenser and evaporator are. (size matters) though often aesthetics plays to much of a roll and we end up with air conditioners with pore COP ratings. Check out some of the best small split systems from Dakin, Mitsubishi. One thing your will soon notice with them is that for a 2 - 2.5kw output rated system the condenser is MASSIVE! a good 2x bigger than a typical (cheaper) less efficient split system of that size and a good 3x larger than a portable system.
This is the reason why as you go larger in your split system starting at say 2kw output and compare it to a 5 or 7kw unit the efficacy drops (almost in 1/2) As the manufactures tend to use the same size condenser unit for several models . Starting at say 2kw and up to 5kw. Then they have one size for all the larger units. Cosmetics, handling a bulky item, shipping all come into play at the expense of efficiency.
You end up being much better off installing two separate 2.5kw units than one 5kw unit from the same manufacturer.
Sorry about that rant on Air conditioners just good info if your looking for a unit.
Happy for the discussion on cooling battery and components to be separated if its to off topic.
Posted: Sun, 29 Mar 2015, 04:39
Today I found some time to hook up the PIP4048 in the power room and to my House 48v 400ah calb bank. I wanted it in place in the power room as a backup to the main Selectronics inverter. Though it will be used to run the Imiev charger (2200w). Just to keep the full 6kw continuous potential of the Selectronics inverter free to run just the house. The aim is to Wear out the cheep inverter with the mundane job of charging electric cars every day.
I was thinking how Weber mentioned he was keen to see his client give the pip system he installed a good workout. load it up a little. So I though while I'm waiting on the change over switch to arrive. I just have the PIP4048 powering one single 15A outlet. I could put his mind to rest on how it handles large spike loads anyhow and test the unit out now it has a solid lithium battery powering it.
The biggest single AC surge load I have is my BOC 17 air compressor. It's a good size 3cyl 17cfm unit with a 2.3kw motor though it spikes way higher than that on start up. Put it this way it makes the lights dim a little in the garage of the On grid house in Brisbane
) It Makes the selectronics grown and shudder a little starting it to offgrid. I don't think their would be a need to start any bigger motors in a typical domestic house application.
I made a small utube video (link below) with the results with the PIP4040 and th air compressor.I should point out that my unit isn't totally stock some caps and fets have been replaced - upgraded though I don't think it would have that much of a impact on it's performance in this test.
https://www.youtube.com/watch?v=QFTZCiG ... e=youtu.be
I tried the unit with a few more loads. A portable Air conditioner, 200A tig welder welding some Aluminium cranked all the way up and then spent the afternoon charging our Imiev at 2200w. Overall I was surprised how it handled the loads.
The unit didn't really get hot. The fans run flat out as soon as any load is applied so it gives you the impression there is a lot of action going on but it's only cold air coming out of the case. The units them self are surprisingly quiet (no buzzing and gowning like the 50hz selectronics) I need to look back though this thread and do the fan modification - replacement that Weber did. 747 jet fan is way over the top and not very smart in it's regulation a bit all of nothing.
Posted: Sun, 29 Mar 2015, 14:41
Replacement caps & fets? Do you guys have a service manual or something? So far there is a dead PIPs & Rpi here, casualties of the learning curve. The Pi is a throwaway but we'd be interested in PIPs resurrection.
Posted: Sun, 29 Mar 2015, 15:00
My unit was brand new and working. Though some parts caps/FETs were upgraded when Weber orderd some parts to repair two broken PIPs for a friend (damaged due to external charger going into runaway voltage)
Posted: Mon, 30 Mar 2015, 17:16
Back to keeping lithium batteries and power electronics cool.....
After spending a few hrs in the power room over the weekend experimenting with the PIP4048. The room is (1.5m x 3m) solid cinder block walls on 3 sides and double barn doors as the front wall with insulated white colorbond roof and concrete floor.
I can say that things start to get warm in there with all the power electronics running at once. There slight inefficiencies combining and showing up as heat in that small room. Two 4000w charge controllers, one 6000w inverter and now an additional 4000w inverter along with some small contributors like the two 200A contactors in the BCU unit that run warmish. My laptop and I guess me.
So yesterday I thought bugger it lets chill the entire room down and keep everything and every one happy. I pulled the pin and ordered a split system Airconditioner for the power room.
Dakin have two models Daikin US7 Ururu Sarara 7 series that are 7 star efficiency but they are silly expensive and apparently more expensive to install as well due to extra fresh air ducting along with high maintenance external filters in dusty environments (I'm on a dirt road). So although they are great. In this application I couldn't justify one.
After a days procrastinating I settled on the Samsung AR5000 Series 2.5/3.2kW Reverse Cycle Inverter Split System Smart Air Conditioner. It uses 500w when cooling vs 420w for the Daikin US7 Ururu Sarara 7 so still not bad. 80w more consumption when you have 8200w pv and 10kw of combined inverter potential isn't a big ask.
The Samsung is wifi enabled. So you can control it via your smart phone or pc. you would view this as a gimmick but for my application it's perfect. This feature is nice for my power room as it's a separate room away from the house and going out there each time to operate the line of sight remote is hardly practical.
It also gives you live feedback via the app of the room temp, advanced timers over the week and can even send you a text message if the unit has a issue. So this should come in handy when I'm not on location or want to stabilize the room temp at anytime off location or on location just from inside the house. Not having to enter the room to use the remote is good in other ways to as it keeps the power room doors sealed shut. This will cut down on dust in the power room being totally seal up.
Price was $820 but ebay have a great deal going until this thursday 2/4/2015 (20% off all electronics) so it was just $656 out of pocket. (The dakin was $2200!)
I ordered a fujitsu 2.5kw split system as well for the clean room in the offgrid shed. Were I am slowly setting up a lab for what I call clean work. Away from welders grease and swarf and so on.
That Aicon was an even better deal $900 but with 20% off so $720 but then fujitsu are offering $150 cash back so it was only $570 out of pocket. It has similar efficacy to the samsung but no wifi control. I think Fujitsu are offering $200 cash back on 5kw or over systems. so very good opportunity if you're in the market for a air conditioner and buy through ebay (I used the good guys ebay store so I could pick the units up myself)
Posted: Mon, 13 Apr 2015, 03:08
I have been testing my PIP4048 for a couple of weeks now as a dedicated Inverter to run my Imiev charger at my offgrid house. It's connected to my 16 cell 400ah calb lifepo4 bank. I have it wired to a single 15A GPO.
Ok so the good and the bad news.
The good news is it handles the steady 2200w load aspect of the I Miev charger without breaking a sweat.
The bad news is it went crazy on me while charging the Imiev yesterday and it did it again today and I got it on video though haven't finished uploading it yet. (will post a link to the video soon)
So what is (going crazy)..... Well As most would know the Imiev charger when first turned on spends about 5 seconds doing a preliminary hand shake - safety check and same other Communication between the EVSE and charger to tell it what rate to charge at. If that checks out ok it slowly ramps up the load (soft start) over the next 5 seconds or so before leveling off at a steady 2200w (9.2A) as it's setpoint for the stock EVSE.
Every now and then, perhaps two or three times Over A full charge the charger will stop for a short period of time lowering the load to nothing (I guess just to check cell voltages balance) before ramping back up again to full output and continuing to charge.
I came out to check on the PIP4048 just as the charger went into a (stop and check) Period. I was eeing a stable 230v - 50hz from the pip and about 5w load. Then as soon as the charger went to slowly ramp back up again the PIP went crazy. The AC output voltage of the pip was bouncing up and down ( 210v - 257v and everything inbetween) It was showing a DC load swinging from (10A - 75A) and the output load was showing (500w - 3300w) Yikes!
I shut it down and Plugged the Imiev EVSE Into my Selectronic Inverter And it just ramped up and started charging at a steady 2200w with a stable 240v AC as expected
So I unplugged the charger from the Selectronic and tried it again in the PIP4048. Again it went crazy as soon as the charger tried to ramp up to max output. (Strange that it handled the ramp up the first time no problem. It has charged the car a few times before this and didn't play up on ramp up) So I switched it off again.
The only difference I had between the selectronics and the PIP4048 was a small 10m 15A extension cord on the PIP4048. So just to eliminate that I pulled the car up closer and plugged the EVSE directly into the 15A GPO. This time the PIP4048 ramped up as normal and started charging.
Could it be that 10m extension cord or just a coincidence. I don't think the extension cord had anything to do with it. In fact I would say I'm sure it wasn't. By this time I had to go pick up my wife from Lansborough train station so that was all the charging I did for the day.
The next day - Sunday 10:00am. I just plugged the Imiev back in. Evse directly to the 15a GPO of the PIP4048. It ramped up as usual and stabilized at 2200w. I checked on it several times over the day as I was working in the area. It charged the car for several hrs from 40% SOC.
The few times the charger paused and ramped back up again the PIP4048 had no trouble and it was business as usual.
Then the trouble started again. This time as the charger was ramping down near the end of charge it went crazy again and repeated the crazy stuff mentioned above
So now I'm back charging with the trusty old Selectronic's as I have charged the car a good 100 times from the Selectronics without issue.
Would be nice to work out what's going on with the PIP4048.
I was thinking a few things. Could it be the PIP4048's output is making the charger or evse signal to the charge do crazy things. Or just the PIP itself going crazy. Either way the PIP is to blame some how.
I seem to remember talk of how a variable output / speed heat gun Making the PIP4048 do something similar?
I know my Selectronics hates my variable heat gun/ paint stripper and that's about the only load I have found that it plays havvoc with it. Though it has no issue at all with the Imiev charger.
Again the strange thing is the PIP had intermittent issues with the Imiev charger ramp up.
Posted: Mon, 13 Apr 2015, 04:40
Posted: Mon, 13 Apr 2015, 05:10
i wonder what would happen if you put a 100W incandescent on the same GPO wile the car is charging would the extra load help stableise it?
Posted: Mon, 13 Apr 2015, 05:33
Fascinating stuff, Kurt. Thanks for posting it. My guess is that it is because both the PIP and the iMiev charger are "smart" i.e. feedback-controlled devices and they get into a state where they are both "chasing each other's tails".
When you run it off the Selectronic, I assume it is not the only load, whereas it is when you run it off the PIP. And I assume it is closer to the PIP's maximum output than the Selectronic's. I wonder if the addition of a small amount of "dumb" resistive load in addition to the iMiev charger might stabilise the PIP. This is not intended as a "solution". Merely as an experiment to gain more insight into the problem. You might then present the data to the PIP's manufacturer. It's possible that just tweaking some PID-controller gain and time-constant parameters in the PIP would stabilise it.
It is hard to know what is going on with infrequently-updating digital displays. You need an analog meter showing AC voltage, or a scope. I'd be happy to loan you the former. And the latter if I accompany it.
It would also be good to see the AC current waveform, for which we'd need a shunt or a current-transformer in addition to the scope. Those variable heat guns can cause problems because their current waveform is highly nonsinusoidal and in the worst case it is asymmetrical between positive and negative swings. But I'd expect the iMiev charger to perform so-called power-factor correction (PFC), i.e. to maintain a sinusoidal current waveform in phase with its voltage waveform.
Posted: Mon, 13 Apr 2015, 05:44
Adverse Effects wrote: i wonder what would happen if you put a 100W incandescent on the same GPO wile the car is charging would the extra load help stableise it?
Indeed, an incandescent bulb would be an excellent dumb resistive load as it would also function like the analog meter and let you see the frequency of the oscillation, if it does not stabilise it.
But does anyone still have such hideously inefficient form of lighting?
I guess you can still get halogens.
Posted: Mon, 13 Apr 2015, 06:01
I did think along the same lines that a small load might help. I plugged in a led workshop light that consumed around 20w but I got the same outcome. Though I can try it with a larger more basic resistive load.
Regarding the imiev charger being close to the max of the PIP4048. The stock EVSE is limited to 9.2A so around 2100w. Just over 50% of max continouse load rating of the PIP.
Yes there is always a good 150 - 200w base load on the selectronic inverter.The selectronic is rated at 6000w continuous.
What was strange is I was registering 3200w at the hight of the fluctuations on the PIP but the EVSE always limits the charger to 9.2A. 3200w is the max of the charger in the I miev but you don't see it unless you plug in a EVSE that tells it to charge at that rate.
The PIP is at the sunshine coast I will have a play with it next weekend.
Posted: Mon, 13 Apr 2015, 07:26
I think Weber may be on the track of wanting to see the waveform from the PIP.
I have viewed the voltage waveform from numerous inverters and gensets with an oscilloscope while charging my i .
I have found the LF pure sine inverters to be the cleanest though often less correct sine shape.
On some HF inverters I have used an inline filter as the i EVSE would reject the power as being too noisy or occasionally shut down.
The filter type is the series inductor / common mode inductor type with caps. It helps reduce the HF components.
BTW, I have never experienced the pause in charging that you describe ? Is it the i unhappy with the power and starting the ramp up again ?
The other chance is that the noise on the inverter waveform is getting through to the pilot signal in the EVSE and the i thinks for a moment that it is being told to charge at full 3.3kW ? Then backing off again. That is, just doing what it is told.
Posted: Mon, 13 Apr 2015, 14:39
"BTW, I have never experienced the pause in charging that you describe ? Is it the i unhappy with the power and starting the ramp up again ?"
I think it's very common. It has this break in charging around 1/4 full (I was mistaken thinking it did it more than once over a charge). On the grid power or Inverter. Many other people with Imievs report the same thing. It then keeps charging. I can dig up posts referring to it from other people. (or I can just capture it on video or canion at the on grid house on mains power)
Edit: dug up the link to pause in charge.
and a graph showing the pause at 3300w.Aparently the pause can last anywhere from 6 - 20 mins
link to post with more graphs showing pause.
"The other chance is that the noise on the inverter waveform is getting through to the pilot signal in the EVSE and the i thinks for a moment that it is being told to charge at full 3.3kW ? Then backing off again. That is, just doing what it is told."
This is what I would place my bet on. Just because when it plays up the charger was ramping between min charge rate and max charge rate (3.3kw) The stock EVSE won't do this unless the pilot signal is telling it to. So I think the PIP is interfering with the pilot signal (sometimes but not every time).
This would explain why it plays up every time the charger needs to ramp up or down. It dosnt know what to ramp up to due to the pilot signal being interfered with and jumps up and down.