I don't really like my Surrettes

Steve Higgins

Re: I don't really like my Surrettes

"So why aren't Outback, MorningStar, Midnite Solar, and Xantrex supporting the Ia / finish stage? I think they should support it. Just put in an option to enable/disable it. "

Cost... Adding a cost of just 10 dollars to an inverter/controller increases the cost of the device 5 to 10 times because of margins.

"Whole bank, open-circuit vs 36-50 amp load (18-25 per string). "

At that current 18-25 amps per string is right around the 15-20 hour discharge rate. a 1.5volt drop from a resting fully charged voltage would'nt be a worry for me.

Steve - from studying our XW with the covers off it is apparent there is no shunts in there. I think they use current transformers to measure the AC side and calculate the DC side in software. At best I think they are maybe +/- about 5% accurate.

I have a friend at Schneider whom I trying to confirm exactly how they are measuring this... my guess like all inverter/controller manufactures is they are measuring ac current, and making a calculation. Yes this would make this measurement really inaccurate.

Steve Higgins

Re: I don't really like my Surrettes

stephendv wrote: »

The bit that's still unclear to me is whether Surrettes can be charged at 2.4Vpc if there was enough time to charge them, or is a high voltage absorb mandatory under all charging and discharging conditions?

In these systems the batteries are only being charged for a short duration of time. Running a bit higher voltage once every two, three, or four days isn't going to cause the damage that running a high voltage into the battery over a long period of time.

A symptom of running too high of a voltage for too long is the SG start to get into the high SG ranges 1.300 and up.

With Surrette Batteries, if you start to see the SG exceeding 1.280 then you simply back off the voltage an and the SG should drop.

YehoshuaAgapao

Re: I don't really like my Surrettes

ChrisOlson wrote: »

boB at midnight solar indicated to me that they will have this feature at some point in the future. I have experimented with it, using Auto EQ in the Classic, to simulate the Ia finish stage. I have sent my results of what it did to boB, and how it works compared to the "boost charging" that XW uses. I did determine that the finish stage using the Auto EQ in the Classic works better than XW's "boost charge" at the front end of the absorb stage.

boB said it can be done just in software and doesn't require any hardware changes. But he's got a lot of irons in the fire and it will not happen right away.
--
Chris

From what I understand, boost charging applies at the transition from bulk to absorption - where bulk goes up to the bulk voltage setting and the bulk voltage is held for one hour (with no current-based exit) before the voltage is dropped to the absorption voltage setting and then absorption exits when current drops to 2% of AH capacity.

Finishing charge is after absorption is complete (56.4-57.6V). When reaching 1.2% of AH capacity (9.6A for 800AH) at the absorption voltage, current is held at 1.2% of AH capacity and voltage is allowed to rise until it hits 2.65 volts per cell (63.6V for 48V bank).

Not sure if finish charging can be 'faked' with Xantrex. Think 'boost charge' (no current-based exit) is activated only when bulk exceeds absorption, not when bulk is less than absorption. I can try it on the weekend - try to set bulk at 57.6V and absorb at 63.6V. It may not work at all, or it may end up feeding 40 amps of current at an elevated voltage, which probably isn't good for the batteries, or the current could be reasonable (under 20A) and achieve something similar to an actual finish charge. Amps from solar is low enough for experimental battery voltage settings until 9AM (30A at 8AM, 40A at 9AM) but after that the amps climb until about 110A at 12:30PM.

ChrisOlson

Re: I don't really like my Surrettes

YehoshuaAgapao wrote: »

Not sure if finish charging can be 'faked' with Xantrex.

I THINK it can by setting the Float stage higher than Absorb? But I've never tried it, so I base that only on the thought that I think XW will let you do that. But just like my experiments with the Classic, it won't be a true Ia finish stage because the finish stage in a real IUIa charger is a constant current stage based on ah capacity.

But assuming the Float can be set higher than Absorb in XW, it's "close enough for government work" to try it. I had really excellent results doing that with the Classic, with the exception that I could not get the AutoEQ to come on when it needed to so the only way to make it work was manually. And I'm not around watching my controllers all the time to manually do it.
--
Chris

YehoshuaAgapao

Re: I don't really like my Surrettes

Another thing unique to my system that is not like everyone else is that I have a grid-tied bi-modal system using batteries designed for off-grid use. I am selling power to the grid during the absorption phase and the inverter and charge controllers have to communicate over the network to tell the charge controllers how many amps the inverter is drawing. This is called enhanced interactive mode - activated by setting grid support voltage equal to equalization voltage (essentially has the charge controllers tell the inverter what grid support voltage should be). Without enhanced interactive mode, the batteries would never get a proper absorption charge off of solar as the inverter would hold the battery voltage at the grid support voltage setting the whole time.

Xantrex is certainly built to favor AGM batteries - Xantrex is built from the old Trace inverters that grew out of the California power crisis and every drawing I saw originating from California included Concorde Sun Xtender AGM batteries. AGM batteries cost twice as much and last half as long and don't tolerate high temperatures very well. Since Xantrex supports peak load shaving (also originating from California where utilities charge extra if peak demand goes over a certain KW level during daytime hours) I'm using that to give the batteries the cycles they need. Late spring / early summer to hoard on-peak credits 5 days a week for A/C use in July/August (swamp coolers rest of time). Fall and early spring one day per week for cycle exercise. December/January can have off-grid days (these are more fun than load shaving) since days are short and won't be wasting solar production.

ChrisOlson

Re: I don't really like my Surrettes

YehoshuaAgapao wrote: »

Since Xantrex supports peak load shaving (also originating from California where utilities charge extra if peak demand goes over a certain KW level during daytime hours)

Whenever I read or hear about folks doing all these things with their grid-tie stuff I set back and breathe a sigh of relief, happy that I don't ever have to deal with any of that. On my side of the fence, inverters are built with dual AC inputs so you can hook two generators up to 'em

YehoshuaAgapao

Re: I don't really like my Surrettes

ChrisOlson wrote: »

Whenever I read or hear about folks doing all these things with their grid-tie stuff I set back and breathe a sigh of relief, happy that I don't ever have to deal with any of that. On my side of the fence, inverters are built with dual AC inputs so you can hook two generators up to 'em

I'm guessing you have the Honda on AC2 which you use for more often for heavier loads with generator support while the Champion is on AC1 for the occasional need of catch-up bulk battery charging.

YehoshuaAgapao

Re: I don't really like my Surrettes

Steve Higgins wrote: »

Cost... Adding a cost of just 10 dollars to an inverter/controller increases the cost of the device 5 to 10 times because of margins.

I think some people, especially people with forklift batteries (would probably also include Trojan Industrial and Surrette 4500 and 5000 series) would fork up the extra $50-100 ($10 extra production cost times 5-10) for an industrial grade IUIa solar MPPT charge controller. If margins are that high, how you get these things directly from the slave factory in China so they could be had for $100 or less? Alliexpress and alibaba (and ebay to a lesser extent, one off-shore middleman) lets consumers order direct from the slave factories but a lot of that stuff looks like junk like Mao-era steel.

Steve Higgins wrote: »

At that current 18-25 amps per string is right around the 15-20 hour discharge rate. a 1.5volt drop from a resting fully charged voltage would'nt be a worry for me.

Its actually measuring when coming out of load shave. Generally from 47.9-48.2V at close to 8PM settling up to 49.4-49.6V at about 8:30PM. When load shave starts it goes down very quick from 51.2 (temperature-corrected float - 1V) to 49.6-49.7V and quick but less so to about 48.8V then is stable from there.

ChrisOlson

Re: I don't really like my Surrettes

YehoshuaAgapao wrote: »

I'm guessing you have the Honda on AC2 which you use for more often for heavier loads with generator support while the Champion is on AC1 for the occasional need of catch-up bulk battery charging.

No, the Champ is really too little to use for battery charging. We just have it hooked to AC1 for backup in case something happens to the Honda. But I still run it from time to time when loads are light to exercise it.
--
Chris

boB

Re: I don't really like my Surrettes

stephendv wrote: »

It's not that simple, because the final "I" stage has to be current limited to no more than 5A/100Ah. But charge controllers have no way of knowing how much current is actually going into the battery, so if they apply a current limit, and then you turn on a heavy load you'll start drawing from the battery because the charge controller has limited its output. Midnite have been promising an external shunt/battery monitor which will make this possible, so hopefully one day we'll have this feature.

Yep. This should be pretty easy once the WhizBang Junior battery monitor is done and shipping.

Also, the boost mode will want to use this meter as well as helping to determine the ending amps.

If this works better, and it is sounding like it does, this will be a no brainer. I will be done.

boB

PS. Hi Steve H. !!!

northerner

Re: I don't really like my Surrettes

boB wrote: »

Yep. This should be pretty easy once the WhizBang Junior battery monitor is done and shipping.

Also, the boost mode will want to use this meter as well as helping to determine the ending amps.

If this works better, and it is sounding like it does, this will be a no brainer. I will be done.

boB

PS. Hi Steve H. !!!

Couldn't the same thing be accomplished with a controller that has a learning capability? Battery voltage will change with any load put on it, and periodic calibration could be accomplished by running the charge through the batteries without a load, could it not? Perhaps in the future, the charge controller would get to know the battery (and adapt to it as the battery ages) and figure out exactly what it needs. Or is that getting too complicated?

inetdog

Re: I don't really like my Surrettes

Steve Higgins wrote: »

Does the Xantrex Meter use a shunt to measure current? from what I seeing their meters are voltage based and not as accurate as you would want... How does the Xantrex meter determine end amps?

Let's try to keep this well specified to avoid thrashing and confusion.
At issue are the Xantrex XW or earlier inverter/chargers which may use AC input power to calculate charging current. Don't know for sure.
And the Xantrex C60, etc. Solar CCs. They uses output amps at DC from the CC terminals to determine battery amps. It does not have a shunt, so it cannot distinguish battery charge current from load current in Bulk or Absorb. The C series asks for size of battery in AH and uses that info to calculate the EndAmp value to use for Absorb stage termination when Absorb time is not set. Not well, if at all, documented by Xantrex. Hence to fudge EndAmps to allow for some load current, people but in a bogus AH value.

The Xantrex Meter is only going to be involved in charging decisions if it communicates with inverter/charge and/or CC over Xanbus.

ChrisOlson

Re: I don't really like my Surrettes

inetdog wrote: »

And the Xantrex C60, etc. Solar CCs. They uses output amps at DC from the CC terminals to determine battery amps

What do they use in solar charge controllers to measure DC current out of the controller? I've never been able to figure that out - a hall effect sensor perhaps?
--
Chris

YehoshuaAgapao

Re: I don't really like my Surrettes

inetdog wrote: »

Let's try to keep this well specified to avoid thrashing and confusion.
At issue are the Xantrex XW or earlier inverter/chargers which may use AC input power to calculate charging current. Don't know for sure.
And the Xantrex C60, etc. Solar CCs. They uses output amps at DC from the CC terminals to determine battery amps. It does not have a shunt, so it cannot distinguish battery charge current from load current in Bulk or Absorb. The C series asks for size of battery in AH and uses that info to calculate the EndAmp value to use for Absorb stage termination when Absorb time is not set. Not well, if at all, documented by Xantrex. Hence to fudge EndAmps to allow for some load current, people but in a bogus AH value.

The Xantrex Meter is only going to be involved in charging decisions if it communicates with inverter/charge and/or CC over Xanbus.

Enhanced interactive mode is working well for me - using XANBUS for inverter and CC to work together, CC tells inverter what grid support voltage should be, Inverter tells CC how much load there is. Without enhanced interactive mode and while selling to the grid (max demand - full inverter capacity - always exceed solar), absorption would be for the full max absorption time while the voltage is fixed at grid support voltage (Xantrex recommends float -1V, which is the grid support that is set by the CCs during float, and with Surrette batts, that is 51.5V). Think without grid selling (off-grid), without enhanced interactive mode, assuming solar exceeds demand, would get full voltage absorb for max duration.

inetdog

Re: I don't really like my Surrettes

YehoshuaAgapao wrote: »

Enhanced interactive mode is working well for me - using XANBUS for inverter and CC to work together, CC tells inverter what grid support voltage should be, Inverter tells CC how much load there is....

Thanks for excellent description!
For the less common user cases, where there is DC load on the batteries, neither the inverter nor the CC know about that.

boB

Re: I don't really like my Surrettes

ChrisOlson wrote: »

What do they use in solar charge controllers to measure DC current out of the controller? I've never been able to figure that out - a hall effect sensor perhaps?
--
Chris

The C series uses a PCB trace's resistance as a pseudo-shunt to measure it's output current.
The XW MPPT 60 uses low value resistors.
The Classic uses shunts.

These are all basically "shunts"... Using resistance and voltage drop. Real shunts or resistors change
their resistance very little over temperature range though.

boB

inetdog

Re: I don't really like my Surrettes

boB wrote: »

The C series uses a PCB trace's resistance as a pseudo-shunt to measure it's output current.

Is the trace resistance controlled enough to be useful directly, or does each board or run of boards have to be individually calibrated? Is the calibration done in software during test?

I have seen systems that use laser etching to produce exact value resistances from poorly controlled sources, but that has been with composite resistance material or resistive ink.

--Unbridled curiosity.

vtmaps

Re: I don't really like my Surrettes

Steve Higgins wrote: »

A symptom of running too high of a voltage for too long is the SG start to get into the high SG ranges 1.300 and up.

With Surrette Batteries, if you start to see the SG exceeding 1.280 then you simply back off the voltage an and the SG should drop.

I was under the impression (perhaps wrongly?) that there is a certain amount of sulfur in a cell and that it exists either as sulfuric acid in the electrolyte or as sulfate on the plates. I was also under the impression (perhaps wrongly?) that when the battery is fully charged all the sulfur is in the electrolyte.

Are you saying that when the battery is fully charged (SG = 1.280), there remains sulfate on the plates that can be brought into solution by overcharging (SG = 1.300)?

--vtMaps

NorthGuy

Re: I don't really like my Surrettes

boB wrote: »

The C series uses a PCB trace's resistance as a pseudo-shunt to measure it's output current.
The XW MPPT 60 uses low value resistors.
The Classic uses shunts.

How about XW6048?

waynefromnscanada

Re: I don't really like my Surrettes

Steve Higgins wrote: »

A symptom of running too high of a voltage for too long is the SG start to get into the high SG ranges 1.300 and up.

I've been overcharging my batteries for years and didn't fully realize it until recently. SG of 1.300 and up have been the norm with them for way to long. After more than 10 years they are still working, but acting as if they have way too much internal resistance. Even moderate loads now easily drop the voltage to 11.2, and as soon as the load is removed, the voltage is right back to normal, and SG remains good. These are the same loads that a two or three years ago, would have hardly altered the voltage at all.
Obviously I'm still learning, even after all these years. But perhaps if I live long enough to think I know it all, that will be the time to hang up the proverbial keys.
I can now see a six pack of Surrette S-1380 in the not too distant future. Hopefully through experience, and with much thanks to the good folks on this forum, I've learned enough to properly nurture them so they too will last at least ten years.

Blackcherry04

Re: I don't really like my Surrettes

waynefromnscanada wrote: »

I've been overcharging my batteries for years and didn't fully realize it until recently. SG of 1.300 and up have been the norm with them for way to long. After more than 10 years they are still working, but acting as if they have way too much internal resistance. Even moderate loads now easily drop the voltage to 11.2, and as soon as the load is removed, the voltage is right back to normal, and SG remains good. These are the same loads that a two or three years ago, would have hardly altered the voltage at all.
Obviously I'm still learning, even after all these years. But perhaps if I live long enough to think I know it all, that will be the time to hang up the proverbial keys.
I can now see a six pack of Surrette S-1380 in the not too distant future. Hopefully through experience, and with much thanks to the good folks on this forum, I've learned enough to properly nurture them so they too will last at least ten years.

That all depends on what the term of " Working " and " Last " means to you, They could probably show some voltage after 20 year's if you kept a charge on them and they don't short out.

YehoshuaAgapao

Re: I don't really like my Surrettes

waynefromnscanada wrote: »

I've been overcharging my batteries for years and didn't fully realize it until recently. SG of 1.300 and up have been the norm with them for way to long. After more than 10 years they are still working, but acting as if they have way too much internal resistance. Even moderate loads now easily drop the voltage to 11.2, and as soon as the load is removed, the voltage is right back to normal, and SG remains good. These are the same loads that a two or three years ago, would have hardly altered the voltage at all.
Obviously I'm still learning, even after all these years. But perhaps if I live long enough to think I know it all, that will be the time to hang up the proverbial keys.
I can now see a six pack of Surrette S-1380 in the not too distant future. Hopefully through experience, and with much thanks to the good folks on this forum, I've learned enough to properly nurture them so they too will last at least ten years.

If you are getting surrettes again, I'd wait for the 4500 series (S-1400EX or S-500EX). Series 5000 plates in a series 4000 L16 container. There seems to be a lot of raving good reviews about Crown forklift batteries too.

ChrisOlson

Re: I don't really like my Surrettes

NorthGuy wrote: »

How about XW6048?

That's what I was saying when Steve asked. I've looked inside both covers on our XW and I don't see no shunts in there. There's current transformers, and Schneider tech support told me that the XW measures DC current in and AC current out. The AC in current / voltage is derived. And he also said the readings at low levels are also off due to the relatively high full scale values involved.

I've had my Snap-On meters on the inverter and the DC current value at low loads is about +/- 5% or so. At higher loads it gets pretty accurate, better than +/- 2%. But that leads me to believe that when XW calculates Ending Amps (2% of ah capacity that you have entered in the menu) that it's not real precise.

And how does XW handle this when loads are on? Say you have a couple XW-MPPT60's in absorb stage and the controllers are putting out 100 amps but the inverter has 80 amp DC load on a 1,000 ah battery bank. First of all, do these components communicate with each other over Xanbus so the whole system "knows" there's really only 20 amps going to the battery? So even though the MPPT60's are putting out 50 amps each, they should "know" that we're at 2% of the bank capacity so it's time to exit the absorb stage even though the timer hasn't expired?
--
Chris

waynefromnscanada

Re: I don't really like my Surrettes

Blackcherry04 wrote: »

That all depends on what the term of " Working " and " Last " means to you, They could probably show some voltage after 20 year's if you kept a charge on them and they don't short out.

Well for me, "working" means they continued to provide power after being fully charged, at a level that was acceptable for my needs, and likewise that they last (continue to operate over time measured in years), at a level acceptable for my needs.
They're still OK for light loads, but anything beyond that, the voltage now drops to unacceptable levels while under load. Therefore the end of what for me was a useful life, is clearly (to me), in sight.

NorthGuy

Re: I don't really like my Surrettes

ChrisOlson wrote: »

Schneider tech support told me that the XW measures DC current in and AC current out. The AC in current / voltage is derived. And he also said the readings at low levels are also off due to the relatively high full scale values involved.

They told me the same thing, but I have my doubts.

ChrisOlson wrote: »

I've had my Snap-On meters on the inverter and the DC current value at low loads is about +/- 5% or so. At higher loads it gets pretty accurate, better than +/- 2%.

My XW has a consistent error. It always show numbers above Fluke measurements on the shunt, up to 10%! As the load increases, the absolute error (in A) increases too, but relative error (%) decreases. I though this may be related to how the batteries react to the pulsing current - my batteries are very unusual

It would be real nice to know how these measurements are done/calculated, but Xantrex tech support, for whatever reason, didn't want to disclose any details.

ChrisOlson wrote: »

But that leads me to believe that when XW calculates Ending Amps (2% of ah capacity that you have entered in the menu) that it's not real precise.

There's some error in mine. As loads get bigger, the amp draw from the battery gets overestimated more, so the calculated amps from the controller get smaller and it exits a little bit earlier that it otherwise would. The error is relatively small, but it could be hours of charging between 14A and 17A ending amps. So, I'm workinkg on the extrnal device to terminate it more precisely.

ChrisOlson wrote: »

And how does XW handle this when loads are on? Say you have a couple XW-MPPT60's in absorb stage and the controllers are putting out 100 amps but the inverter has 80 amp DC load on a 1,000 ah battery bank. First of all, do these components communicate with each other over Xanbus so the whole system "knows" there's really only 20 amps going to the battery? So even though the MPPT60's are putting out 50 amps each, they should "know" that we're at 2% of the bank capacity so it's time to exit the absorb stage even though the timer hasn't expired?

XW6048 sends a message through XanBus 10 times a second advertizing the current that it takes from the batteries. It doesn't do that when SCCs are disconnected. Apparently, SCCs use this to calculate correct ending amps. All works well if you do not have external loads/chargers on the battery.

inetdog

Re: I don't really like my Surrettes

NorthGuy wrote: »

As the load increases, the absolute error (in A) increases too, but relative error (%) decreases. I though this may be related to how the batteries react to the pulsing current - my batteries are very unusual
It would be real nice to know how these measurements are done/calculated, but Xantrex tech support, for whatever reason, didn't want to disclose any details.

We might even call them exceptional!

There has been a discussion of a related error in voltage measurement which may also be part of the problem with your Xantrex.
One family of CCs (Xantrex if I remember correctly) has the option to connect two remote voltage sensing leads directly to the battery posts. That theoretically lets the CC know what the battery voltage is even though the voltage on the output wires is higher by the voltage drop in the wires.
But the circuit measures voltage from the - output to the + sense lead instead of the voltage from the - sense lead to the + sense lead.

If you are using a resistor which does not have one end firmly at ground, the "shunt" voltage readings taken from a measurement at one end only will be too high.

ChrisOlson

Re: I don't really like my Surrettes

NorthGuy wrote: »

XW6048 sends a message through XanBus 10 times a second advertizing the current that it takes from the batteries. It doesn't do that when SCCs are disconnected. Apparently, SCCs use this to calculate correct ending amps. All works well if you do not have external loads/chargers on the battery.

So it really is an integrated system that at least does SOME compensation for loading on the inverter during charging. I only ran my XW-MPPT60 for a couple weeks so I never really got to try all that out. The main problem I had with different brands of controllers was getting the MPPT60 to agree with the Classics on what the charge stage should be. Since I got more money in Classics than I had in the XW controller I just removed it from the system.

But that is nice to know - thanks.
--
Chris

Cariboocoot

Re: I don't really like my Surrettes

Morningstar is known for their "battery sense" terminals to give non-current affected Voltage readings. The way they use those readings is to my mind odd, but 'A' for effort.

It may surprise people to know that in general charge controller and inverter meter readings are not exactly "instrument accurate". This is because they don't really need to be; they only need to be "practically functional". Hence there usually is a "sweet spot" where they'll agree with test equipment and outside of that they'll be wrong one way or the other.

It's all about that oh-so-necessary operating tolerance.

boB

Re: I don't really like my Surrettes

As far as I have seen, all of the typical battery based inverters (Magnum, OB, Xantrex, SMA) use AC current
transformers to measure current and derive the DC charging and inverting current from that.

As for the C series CC's, those don't do end amps AFAIK and the copper traces
are only used for over-current protection and ball-park current display.
Just timers to end Absorb.

Copper is not temperature stable so at higher current, will heat up and
show even higher DC current if not somewhat temperature compensated.
For over-current, calibration doesn't really matter much.

boB

ChrisOlson

Re: I don't really like my Surrettes

boB wrote: »

As far as I have seen, all of the typical battery based inverters (Magnum, OB, Xantrex, SMA) use AC current
transformers to measure current and derive the DC charging and inverting current from that.

Schneider told me, quote, "the XW inverter measures DC current in and AC current out. The AC in current is derived."

I seen the CT's in there when I had the covers off that measures the L1 and L2 to loads. But for the life of me I could not see or figure out how they're measuring the DC side. There's some freakin' big stuff in there hooked up to those DC power studs. But I didn't see no shunts in there.
--
Chris