Alternator related links

[markopolo]

Interesting to see the output drop quickly in this video:

https://www.youtube.com/watch?v=iQ0xryKSpEg

The voltage output in the video starts off in the 14.8v range and about a minute later it's has dropped approx 1 volt. Amps at the start were around 200A and about a minute later down to 182A or so.

14.8V x 200A = 2960 Watts
13.8V x 182A = 2512 Watts

Approx 450 watt drop caused by heat I'd assume. The video gives a good idea of what to expect.

---------------------------------------------------------

Some Ram alternator output curves here:
http://www.rambodybuilder.com/2015/d...maltoutcrv.pdf
Pulley ratios look to be 2.72 to 2.9 for gas engines. It does not appear to be Promaster specific but still gives an idea of what they do.

---------------------------------------------------------

A 50A DC to DC charger here - 600 Watt 12 volt to 12 volt DC DC battery charger, Heavy duty and military grade 12VDC to 12VDC DC-input lead acid battery chargers 50A amp charge rate from PowerStream. - connected to the OEM alternator could allow you to get an extra 600W in Class B's that don't utilize the OEM alternator for coach power (B's with a second alternator for coach power). Expensive though.....

----------------------------------------------------------

Ford Transit alternator curves - https://www.fleet.ford.com/truckbbas..._BEMM_v1-0.pdf - check pages 79 through 87. There's temperature data, voltage, rpm. Pulley ratios appear to be 2.7 for gas engines and 2.69 for the diesel.

[booster]

Good information! I have been trying to get a feel for how much reduction there is in alternators at high load, and also how much is really needed for various applications.

Those numbers are actually quite good compared to a lot of what I have found, particularly for big alternators running on the Baldor regulators, which seem to recover in the 60-70% of alternator capacity once hot. The Baldor reduces output to 50% once the alternator gets to 225*F, so it seems to be a bit more conservative than the internal regulators in most internally regulated alternators. For instance, our 250 amp DC Power alternator claims a 3 step temperature compensation, but we did not see much at all on our full load tests, as it started at 200 amps, and ended at about 188 amps over the tests. Unfortunately, we did not have accurate alternator temps at that time. I have heard that DC Power likes to have the case under 270*F, though, so that is a lot more than the Baldor allows. The big question to me is if you are better off to run consistently a bit hotter, but not heat cycle, or not. In most equipment I have dealt with, heat cycling is considered a worst case scenario, so I wonder.

If setup to the normal Baldor parameters, it appears that most of the engine generator setups would not come close to producing the recovery rates that are being claimed based on the max output of the alternator used. More likely, they would be in the 2/3 of capacity range. Using the alternator internal regulator style might give more actual recovery, but you would give up battery temp compensation, so there are downsides to that, also.

[avanti]

Originally Posted by booster

The Baldor reduces output to 50% once the alternator gets to 225*F, so it seems to be a bit more conservative than the internal regulators in most internally regulated alternators.

The setpoint temperature at which the Balmar regulator reduces power is fully programmable, so you can pick your poison. it doesn't appear that you can control the fallback output percentage, but I'm not sure how much that matters. Reducing the percentage to a greater degree simply causes the cooling to happen faster. It is not obvious to me whether the net amps delivered changes much one way or the other.

[booster]

Originally Posted by avanti

The setpoint temperature at which the Balmar regulator reduces power is fully programmable, so you can pick your poison. it doesn't appear that you can control the fallback output percentage, but I'm not sure how much that matters. Reducing the percentage to a greater degree simply causes the cooling to happen faster. It is not obvious to me whether the net amps delivered changes much one way or the other.

Heat is heat, so I would expect the net amp hours recovered would not change much. I think, for me, the question is whether or not to turn down the output to the average (which you can do with the Balmar) to keep the unit from thermal cycling. I do think that raising the temperature that the reduction happens would give more average output amp hours, as the hysteresis is likely at a fixed amount of degrees, so both cut and increase would go up in temp. The 225* may have been chosen to cover stock style alternators that don't handle the heat as well as the high output ones like the DC Power units. From our tests, it appears that DC Power lets them get quite a bit hotter than 225* before turning down the output, when on the internal regulator they supply. If this is the better or worse, is anyone's guess, as nobody that I have found publishes anything on it.

[markopolo]

It appears heat becomes a factor immediately if the amp flow is high. It was a 270A with 180A or so throughput in the video.

To me, it looked like full voltage was back when the load was around 100A. I think the amplifier was turned off then but lights and stuff were still running for a short period. It would be interesting to see the same setup under a variety of loads so we could determine what I'd describe as peak charging efficiency.

Voltage is a key factor for fast battery charging. You'd need to be at a full charging voltage to have the batteries accept the most current.

[booster]

I just watched that video again, actually about 10 times, and what we are seeing might not be what it first appears.

The alternator is interesting in that it is a cast case, not the machined ones like ours is for an XP, but it is stated to be a 270XP and at idle speed. They don't give alternator rpm though. It also is running on an internal regulator, and in an XP, that is probably a Denso style, which is what we do have. It looks from the voltage to be the higher voltage model regulator.

The Denso regulator does some really funny looking things as it runs, at least it does on ours. I have mentioned some of them in other threads when the have come up.

When first started, it will go high voltage, near 15 volts for the high voltage version, then settle to 14.7/14.8v after a short time. The video doesn't show the startup it appears.

The fast drop of the current in the first minute or two does show how quickly they lose output with heat, but there may be something else going on at the same time. You could see the voltage bouncing back up repeatedly in the video, with the current not changing all that much. The bounces were of very short duration, and relatively consistent. We have seen the same bouncing on ours, but not as quickly repeating. I was told the that the Denso regulator is a "smart" regulator that will turn itself down when the batteries get full, and we have seen that happen on ours. I think that the way it controls the turn down is by reducing the output voltage setting and then checking to see how long it takes for the actual circuit voltage to drop. That length of time would indicate how full the battery is and if the power is going to the battery or to loads. It then decides whether to cut back or not. This wouldn't have much to do with the final one volt drop, though, as the alternator was maxed out at that speed. Based on our XP, his reduction probably wouldn't get much more unless he really increased output, as we were able to run with about the same reduction for a long time. It is likely that if he had a factory set Balmar with alternator temp, it probably would have turned down 50% and then cycled, killing the battery quickly.

At the beginning of the video, he lists all the stuff he is running, and it looks as if there may only be one 70ah AGM in the system. The alternator is maxed out, so the voltage drop is probably from the battery actually being discharged, allowing the system voltage to drop. This would make the system much different than a battery charging situation because big loads like that will usually increase their amperage as system voltage drops (like our inverters do). His meter is on the alternator output, but it sure would be interesting to see what the battery amps would be, and which way. His load may also have been highly variable, as the most power would be going to the big woofers I think, and totally dependent on the amount of bass being sent to them.

If he really has only 70ah of battery, he will need to be very careful not to overheat it charging off the 270 amp alternator.

I have done some testing on the the voltage vs charging speed and was a bit surprised with what I saw. On a 50% discharge on 440h bank, I ran it once at 14.3v and once at 14.6v with the 100 amp shore charger. The 100 amps is not enough to hold absorption voltage at 50% SOC, so there was a bulk stage at lower voltages until it climbed to absorption. As it turned out, the absorption stage was not all the long, time wise, and it was even shorter at 14.6v because it was in bulk longer to get there. By the time it got to 75% SOC, it had started to taper the amps because of the battery acceptance. The overall gains were in minutes for the entire cycle, which was a surprise as I expected a much larger time reduction. 20-30 minutes difference in an 8-10 hour cycle is not huge. Obviously, .3v wasn't enough increase to make a big difference.

The biggest thing is to have lots of amps available, so you can get to absorption voltage and gain the benefits of a higher voltage setting more quickly, plus the extra amps go in besides while getting there. Once the amps start tapering, all get pretty equal. 190 amps of charging off the alternator is not enough to hold setpoint voltage with our 440ah batteries 50% down, so it takes a lot of amps to be able to hold a higher setpoint. I also think that with AGMs, if you hit them hard enough to hold voltage through more of the charging, they will get hot on you. At about 180 amps, we saw temp getting close to 20 degrees up, where at 100 amps it was only 10 degrees up in another test. On a warm day, that can get you warmer than you would like to be.

Of course, there are other opinions out there, by lots of experts. I have read several places that they prefer low amp charging to have a very long bulk stage so the voltage stays below the gassing point and oxidation point as long as possible. Lifeline says the key is to charge quickly on 50% and deeper discharges at least .2C, and they are fine at .4C (.4C is about where I think the batteries start to get too warm) so they are just the opposite.

[markopolo]

Those are all very good points. I forgot about the batteries possibly getting too hot.

It becomes clear that being able to select the alternator output would be beneficial. You can choose what best meets the current (no pun intended) situation.

Here's another interesting link highlighting a major difference between lead acid and LiFePO4: Charge voltage experiments with lithium iron phosphate batteries showing how capacity varies with charge voltage and higher cycle live with lower charge voltage

The table indicates you can get to 99% SOC at 13.6V with LiFePO4 batteries and I'd assume that would be relatively fast. I'm not sure if lead acid could get to 99% SOC if limited to 13.6V. If they could it would take a long time.

[MsNomer]

"Some Ram alternator output curves here:
http://www.rambodybuilder.com/2015/d...maltoutcrv.pdf
Pulley ratios look to be 2.72 to 2.9 for gas engines. It does not appear to be Promaster specific but still gives an idea of what they do."

That appears to be for RAM trucks. Much larger engines than the PM. Don't know how much difference that makes. IIRC, our alternators are 180 with optional 220.

[markopolo]

I searched for a long time for Promaster alternator output curve charts but found nothing

[booster]

Originally Posted by markopolo

Those are all very good points. I forgot about the batteries possibly getting too hot.

It becomes clear that being able to select the alternator output would be beneficial. You can choose what best meets the current (no pun intended) situation.

Here's another interesting link highlighting a major difference between lead acid and LiFePO4: Charge voltage experiments with lithium iron phosphate batteries showing how capacity varies with charge voltage and higher cycle live with lower charge voltage

The table indicates you can get to 99% SOC at 13.6V with LiFePO4 batteries and I'd assume that would be relatively fast. I'm not sure if lead acid could get to 99% SOC if limited to 13.6V. If they could it would take a long time.

I think we saw those, or similar, curves when a member was trying to determine what cutoff voltage to use on his lithium setup. Many of us were surprised at how low the voltage was to get to just under 100% full, and wondering where all the recommendations for mid 14 volts for lithium had come from. Speed wouldn't be an issue in this case, as it would be really hard to have a charger big enough to not be dragged down to the lithium battery voltage.

[gregmchugh]

Originally Posted by markopolo

Those are all very good points. I forgot about the batteries possibly getting too hot.

It becomes clear that being able to select the alternator output would be beneficial. You can choose what best meets the current (no pun intended) situation.

Here's another interesting link highlighting a major difference between lead acid and LiFePO4: Charge voltage experiments with lithium iron phosphate batteries showing how capacity varies with charge voltage and higher cycle live with lower charge voltage

The table indicates you can get to 99% SOC at 13.6V with LiFePO4 batteries and I'd assume that would be relatively fast. I'm not sure if lead acid could get to 99% SOC if limited to 13.6V. If they could it would take a long time.

Looking at this info seems to indicate that if an AGM charge profile is used for a mixed AGM and lithium battery bank with the float voltage 3.4v or higher then the lithiums will get charged to at least 96% SOC...

Or am I missing something?

[booster]

Originally Posted by gregmchugh

Looking at this info seems to indicate that if an AGM charge profile is used for a mixed AGM and lithium battery with the float voltage 3.4v or higher then the lithiums will get charged to at least 96% SOC...

Or am I missing something?

I think you are right on that, except that the lithium would be full long before the AGMs due to their acceptance rate, and we are now being told that you don't want lithium on long charge cycles after they are full.

[gregmchugh]

Originally Posted by booster

I think you are right on that, except that the lithium would be full long before the AGMs due to their acceptance rate, and we are now being told that you don't want lithium on long charge cycles after they are full.

But, I think the Ecotrek BMS will disconnect the charge terminal when the cells are full which takes care of the underhood generator and the solar.

I still haven't figured out how they deal with the shore battery charger since the inverter is connected to the load terminal not the charge terminal...

[booster]

Originally Posted by gregmchugh

But, I think the Ecotrek BMS will disconnect the charge terminal when the cells are full which takes care of the underhood generator and the solar.

I still haven't figured out how they deal with the shore battery charger since the inverter is connected to the load terminal not the charge terminal...

The whole thing gets really weird if you want to do it right. Even if the BMS disconnects the lithiums, how do any of the charge sources know when to go to float for the AGMs , as they are now seeing way different acceptance rates and capacities. IMO, the only way to accurately take care of all the charging is with shunt based monitors for whatever batteries you have, and a programmable controller. Otherwise it always seems to turn into a cluster---.

[gregmchugh]

Originally Posted by booster

The whole thing gets really weird if you want to do it right. Even if the BMS disconnects the lithiums, how do any of the charge sources know when to go to float for the AGMs , as they are now seeing way different acceptance rates and capacities. IMO, the only way to accurately take care of all the charging is with shunt based monitors for whatever batteries you have, and a programmable controller. Otherwise it always seems to turn into a cluster---.

I guess there is really not much advantage to trying make it right vs adequate for the Ecotrek system. If the Ecotreks get near fully charged and the BMS protects the cells from over charge, over discharge, over temp, under temp,etc. and the AGM has a reasonable life then the vast majority of the owners will be satisfied.

An interesting comment I saw somewhere is that the Ecotrek module includes a cooler to prevent over temp of the cells.

[markopolo]

Originally Posted by gregmchugh

Looking at this info seems to indicate that if an AGM charge profile is used for a mixed AGM and lithium battery bank with the float voltage 3.4v or higher then the lithiums will get charged to at least 96% SOC...

Or am I missing something?

If mixing lead acid and LiFePO4 then I would speculate that ratio of lead acid to LiFePO4 greatly influence the choice of charging profile.

Float voltage for lead acid is typically 13.2V. That voltage would only put 20% to 31% back into the LiFePO4 batteries according to the data in that table.

3.4V per cell x 4 = 13.6V The 270 XP alternator didn't dip below that in the video from the first post so it would seem likely that a LiFePO4 bank would easily get full or very near full given enough time idling. Standalone lead acid batteries would also get near full assuming that as the amp flow required from the alternator decreased the voltage then increased to 14.4V or over.

You'd only know that the batteries are full or near full if you know how much current is flowing in. There would be little point in idling for an extra 30 minutes or an hour to get an additional 1% of LiFePO4 capacity. If a coach doesn't have that level of monitoring built in then you could use a clamp on ammeter to observe current flow. It loops over - encircling insulated wires - no disconnecting or using probes etc.

I think the data in the LiFePO4 table indicated charging was terminated when current in dropped to 1.35% (or so) of battery capacity.

[booster]

With Lithium and engine charging, plus an AGM, a better setup might be with a fixed voltage alternator at 13.6v and a small DC to DC charger for the AGM, which probably won't need much charge anyway. Let the BMS shut off the lithiums when they are done. You probably could use the same theory for the shore and solar charging also, using the DC to DC charger on the AGM, and fixed voltage chargers.

There is one other issue, though, if the system did not have an AGM, and is probably a secondary reason (after recovery) that Roadtrek uses one, and that is that you would need to be sure that the BMS turned off the alternator itself, preferably a touch before the batteries disconnect, so the alternator doesn't self destruct.

[markopolo]

Would LiFePO4 accept higher amps when the additional force of higher voltage is applied? I'm assuming that is how it works just like lead acid. If so, then limiting voltage below 14.4v or 14.6V could slow the charging down. I know it's unlikely anyone would install enough alternator output to satisfy what larger LiFePO4battery banks could accept.

[gregmchugh]

Originally Posted by markopolo

If mixing lead acid and LiFePO4 then I would speculate that ratio of lead acid to LiFePO4 greatly influence the choice of charging profile.

Float voltage for lead acid is typically 13.2V. That voltage would only put 20% to 31% back into the LiFePO4 batteries according to the data in that table.

3.4V per cell x 4 = 13.6V The 270 XP alternator didn't dip below that in the video from the first post so it would seem likely that a LiFePO4 bank would easily get full or very near full given enough time idling. Standalone lead acid batteries would also get near full assuming that as the amp flow required from the alternator decreased the voltage then increased to 14.4V or over.

You'd only know that the batteries are full or near full if you know how much current is flowing in. There would be little point in idling for an extra 30 minutes or an hour to get an additional 1% of LiFePO4 capacity. If a coach doesn't have that level of monitoring built in then you could use a clamp on ammeter to observe current flow. It loops over - encircling insulated wires - no disconnecting or using probes etc.

I think the data in the LiFePO4 table indicated charging was terminated when current in dropped to 1.35% (or so) of battery capacity.

I took a look at the standard float voltage for the AGM profiles on the three chargers (solar, inverter, and aux alternator) that are used in an Ecotrek van and all have a Float voltage of 13.4 volts so if the Ecotreks were not fully charged when the charger went into float they would get to 96% charge which I think is close enough for their typical usage scenario.

[booster]

Originally Posted by markopolo

Would LiFePO4 accept higher amps when the additional force of higher voltage is applied? I'm assuming that is how it works just like lead acid. If so, then limiting voltage below 14.4v or 14.6V could slow the charging down. I know it's unlikely anyone would install enough alternator output to satisfy what larger LiFePO4battery banks could accept.

I think the only test I have seen that had enough charger to hold charger voltage was on very small batteries. All the charge curves I recall seeing for RV stuff climbed the voltage, at essentially constant current, until it got to charger setting. We did see some that were 14.6v chargers, where they wanted to stop at a bit under full charge on the lithiums, and did it by stopping based on voltage of something like 13.6v, so they never got to use the higher voltage.

It would be interesting if someone that has a lithium bank and a big engine generator could look at if the amps are enough to get the charge voltage to get higher than the battery voltage. Simple test, I think, as all you would need to do would be to see what the voltage was while charging near the end of cycle and then stop the charger and see what the batteries settle at. The further those voltages are apart, the more the higher voltage was gaining you, I think.

I think the next time I do a full recharge from 50% or so, with the shore charger, I will try to get a real number on the charger rate difference between 14.3v and 14.6v on our AGMs. If I just turn up the absorption setting, once it is in absorption, I should see an amps increase. My guess is that it probably will be a bit high initially because it will add a little surface charge, but would likely settle pretty quickly to a new stable rate that is slowly reducing. I have done that in float to see how much difference there was, and also right at the end of absorption to see how much voltage altered the float transition amps setting. At those very low amps, the amp changes were small numbers, but fairly large as a percent.