Cruisers Forum
 


Join CruisersForum Today

Reply
 
Thread Tools Rate Thread Display Modes
Old 15-07-2011, 02:20   #1
Registered User

Join Date: Dec 2010
Location: Sydney
Boat: Beneteau Oceanis 46
Posts: 47
Xantrex Link Pro Overestimating SOC Percentage

Hi All,

I have just installed a Xantrex Link Pro in an attempt to better understand my state of charge. I have a 720 amp hour battery bank that I suspect was damaged by the previous owner.*

The Link Pro has been professionally installed and I've confirmed the basic function settings and it's synchronized but the readings dramatically overstate the remaining capacity. For example the SOC% may read 90% when the resting volts are reading just 12.1V.

What I have found is that the SOC% display is no more that the amp hours used divided but the total capacity set in the function menu. For example if the amp hours consumed as counted by the link pro was 72 it displays 90% for my 720 battery bank irrespective of the voltage. I thought the link pro was smarter than this?

So as I've said I suspect my batteries capacity is no longer 720, I've tried fooling the link pro by changing the total capacity to a lower figure of 400 and the SOC% is a bit more normal when compared to voltage readings. Is it correct to do this?

Finally I'm interested in doing a proper capacity test to find an exact figure for my bank, does anyone have any good information on how to do an accurate capacity test? Will a capacity test further damage the bank? I read once that fully discharging can even bring back some life into them?
__________________

__________________
Hugh Walker is offline   Reply With Quote
Old 15-07-2011, 06:31   #2
Registered User
 
Stu Jackson's Avatar

Join Date: Jan 2011
Location: Cowichan Bay, BC (Maple Bay Marina)
Posts: 6,385
Re: Xantrex Link Pro overestimating SOC%

You might be interested in this for your Link: Link-series Charging Algorithms -- The "Gotcha" Factor!

The Link 2000 manual has a capacity test description.
__________________

__________________
Stu Jackson is online now   Reply With Quote
Old 28-07-2011, 02:29   #3
Registered User

Join Date: Dec 2010
Location: Sydney
Boat: Beneteau Oceanis 46
Posts: 47
Re: Xantrex Link Pro Overestimating SOC Percentage

Hi Stu,

Thanks but I can't open the link, the website keeps saying an error has occurred. Are you able to post the text?

Cheers,

Hugh
__________________
Hugh Walker is offline   Reply With Quote
Old 28-07-2011, 06:37   #4
Registered User
 
GeoPowers's Avatar

Join Date: Dec 2008
Location: Gulfport, MS
Boat: Beneteau 393
Posts: 947
Images: 27
Re: Xantrex Link Pro Overestimating SOC Percentage

Quote:
Originally Posted by Hugh Walker View Post
Hi All,


What I have found is that the SOC% display is no more that the amp hours used divided but the total capacity set in the function menu. For example if the amp hours consumed as counted by the link pro was 72 it displays 90% for my 720 battery bank irrespective of the voltage. I thought the link pro was smarter than this?

So as I've said I suspect my batteries capacity is no longer 720, I've tried fooling the link pro by changing the total capacity to a lower figure of 400 and the SOC% is a bit more normal when compared to voltage readings. Is it correct to do this?

Finally I'm interested in doing a proper capacity test to find an exact figure for my bank, does anyone have any good information on how to do an accurate capacity test? Will a capacity test further damage the bank? I read once that fully discharging can even bring back some life into them?
Hi Hugh,

I can't comment on the Link system, but my Magnum battery monitor computes SOC % the same way, by taking the Ah capacity you plugged in and dividing by what it saw passing by the shunt.

My 454 AH house bank batteries are 5-6 years old, and definitely don't have 454 AH capacity any more. In order to get a better guage of the capacity of this bank I ensured a full charge dockside, reset the Ah I/O reading to zero, and then shut off the charger. I let it discharge overnight and into the next day and then turned the charger back on. Then what I did was I continually monitored the charge display and assumed an 80% SOC when the charger shifted from bulk to aboprtion phases. I looked at the display of +/- of Ah consumed. I multiplied that number by 5 and that was my new house bank capacity. It isn't exact but it is accurate enough for me and more accurate than just going off the label of some old batteries.

Hope this helps,

Frank
GeoPowers is offline   Reply With Quote
Old 29-07-2011, 10:34   #5
Registered User
 
Stu Jackson's Avatar

Join Date: Jan 2011
Location: Cowichan Bay, BC (Maple Bay Marina)
Posts: 6,385
Re: Xantrex Link Pro Overestimating SOC Percentage

Quote:
Originally Posted by Hugh Walker View Post
Hi Stu,

Thanks but I can't open the link, the website keeps saying an error has occurred. Are you able to post the text?
Hugh,

Here's the first part, which is the basis of it. There are a few more posts on that topic that get into more details. I emailed you via this board, i you want the rest.

********************
********************

Link Algorithm Operation

In discussions over the course of a few years, and since I finally installed my Link 2000 last year, I have been investigating the operation of the Link algorithm and how it affects the indication of "complete" recharging.

Last year I was involved in a detailed discussion over on co.com about how they work, related to another issue altogether.

Rich Stidger with his Hunter 40.5 from Rhode Island, and Donalex from the U.K., helped me to understand this. In a post today, Rich summed it up pretty well, as follows:

The Link 20 has an "gotcha" that isn't real obvious. When charging, the Link 20 considers the battery to be fully charged when two conditions are met. First the charging voltage must be at or above the entered parameter and the charging current must be below the percentage of the bank capacity that is also an entered parameter. In my case those parameters were 13.2V and 2%. 2% is about 9.2A for my bank of 460Ah. When these two parameters are met, the Link 20 blinks the top green LED indicating full charge, resets the AH used to zero, and recalculates and stores the charge efficiency factor for future charging. This can happen even if the used AH are not counted down to zero.

Without realizing this was happening, my golf carts were being cheated of 10-20Ah at various recharge cycles during the month. Thus I thought my batteries were being fully recharged but actually they were reported to be full but were short. Over the month I calculated that they were about 140Ah down out of the 460Ah capacity. Thus I had a chronic undercharge.

This undercharge showed itself as a low terminal voltage in the morning, sometimes as low as 11.9V when the voltage was 12.4-12.5 or so when retiring for the night.

The scenario for the Link 20 gotcha is this: The charging current has tapered down to about 13A with 20Ah still to go with the charging voltage at 14.4V. Now one of the refrigeration units turns on drawing 5A. This causes the battery charge current to drop below the 2% threshold and after 6 minutes the Link 20 thinks the batteries are fully charged because the charge voltage is over 13.2V, the charge current is below 2% or 9.2A, and these conditions have been met for 6 minutes. Then to make matters worse, the recharge efficiency factor is recalculated to a higher value (ex. 0.97 vs 0.95) so the at the next charge cycle the Link 20 thinks that the batteries require fewer input AH for the used AH. This is not true of course, but it makes the under-charge condition worse.

The fix for this gotcha is to set the Link 20 parameters to 15v and 1%. These parameters are not ever going to be met, so that means that the Link 20 will never think the batteries are fully charged until the AH used are fully counted down to zero. Also set the recharge efficiency factor to manual of 0.94 instead of the automatic setting of 0.95. This will stop the automatic recalculation and fix the recharge efficiency to be 0.94.


This "feature" is clearly documented in the instruction manuals, but not specifically noted to end up with this result. I had mentioned it somewhere before here on the MB but haven't put my hands on it yet (oh, no!). It's another one of these "do-the-math" things because when the bank is close to charged and the fridge kicks in this happens all the time. I tried it and tested it and it works. I haven't changed the parameters on my unit yet, but have these kind of notes in my on board Link manual and in my Electrical system notebook.

[Added July 14, 2009] I have changed these parameters on our Link and the results have been fantastic and exactly as Rich described. Before, our Freedom 15 Inverter/charger combo, after a daysail, would go from bulk to absorption to float within ten minutes of being plugged in. After the change to the parameters, the charger stays on absorption for a much longer time. Ron Hill's comment, below, is also another way to check if you do the charger controls manually. Since our Link controls the charger, it is necessary for our use.

Rich is on a mooring and uses generator power and his alternator for charging and is rarely plugged in for extended periods of time. That "gotcha" got him, and, while you think this may NOT get those of you who have access to plug-ins on a regular basis, you're wrong! This is because the Link will be incorrectly reporting full banks when they are NOT full. This is very important. You need to trust your instruments. Knowing how they work will help you. Rich was was simply trying to minimize his carbon footprint and learned about this "the hard way."

As far as I know, this "gotcha" is true for all Link units, because they use the same algorithms. Check your owner's manual for your own particular units. Connected Link "X000" series units with Freedom inverter chargers are particularly susceptible to this issue because the Link 2000 controls the Freedom charger! "Smaller" independent Link units with separate chargers are done "manually" by the skippers by seeing the Link reporting full and operating their chargers accordingly.

Some of this information is relevant to the "Acceptance" thread I began a while ago. Battery Acceptance & Link 20 / Monitor Operation
__________________
Stu Jackson is online now   Reply With Quote
Old 29-07-2011, 11:02   #6
Marine Service Provider
 
Maine Sail's Avatar

Join Date: Jan 2007
Location: Maine
Boat: CS-36T - Cupecoy
Posts: 3,060
Re: Xantrex Link Pro Overestimating SOC Percentage

Quote:
Originally Posted by Hugh Walker View Post
Hi All,

I have just installed a Xantrex Link Pro in an attempt to better understand my state of charge. I have a 720 amp hour battery bank that I suspect was damaged by the previous owner.*

The Link Pro has been professionally installed and I've confirmed the basic function settings and it's synchronized but the readings dramatically overstate the remaining capacity. For example the SOC% may read 90% when the resting volts are reading just 12.1V.
How are you determining "resting voltage"? Has the bank been allowed to rest completely disconnected from any loads or charging sources for 12 hours or more?

Are you sure you have not plugged in a reserve capacity rating vs. a 20 hour Ah rating? I see this one quite a bit?

What are the batteries; brand, model etc.?

"Professionally installed' means little these days and battery monitors are one item even many pros mess up. You may have a load not being counted or captured by the shunt. This is the biggest installation error I come across.



Quote:
Originally Posted by Hugh Walker View Post
What I have found is that the SOC% display is no more that the amp hours used divided but the total capacity set in the function menu. For example if the amp hours consumed as counted by the link pro was 72 it displays 90% for my 720 battery bank irrespective of the voltage. I thought the link pro was smarter than this?

Actually the SOC% screen is Peukert compensated thought the instructions are HORRIBLE and it is nearly impossible to garner this info from the user manual. The Ah screen is your Ah used and your SOC% screen is Ah's used calculated with Peukert. I actuall spoke with TBS Electronics, they actually make this product, about this because nobody at Xantrex had a clue...


Quote:
Originally Posted by Hugh Walker View Post
So as I've said I suspect my batteries capacity is no longer 720, I've tried fooling the link pro by changing the total capacity to a lower figure of 400 and the SOC% is a bit more normal when compared to voltage readings. Is it correct to do this?
This is very likely the case. The Link Pro or any other monitor is only as smart as you are at programming it. If you've lost capacity you need to measure your batts and plug that number in.

Quote:
Originally Posted by Hugh Walker View Post
Finally I'm interested in doing a proper capacity test to find an exact figure for my bank, does anyone have any good information on how to do an accurate capacity test? Will a capacity test further damage the bank? I read once that fully discharging can even bring back some life into them?
The best test for a monitor calibration is a 20 hour Ah test. Take the manufacturers Ah rating of your battery, not the entire bank, and divide by 20. So a 100 Ah battery would need a 5A load for the test. Test the batts individually as one may be worse than the others. You should start with a specific gravity test first though, if flooded...

Find a load that is equal to your batteries Ah rating spec load and use that load for the test. You may need a variable dimmer to dial in the load.

Fully charge the battery then connect the load and monitor it for 20 hours or until it hits 10.5 volts. If it hits 10.5 volts before 20 hours you need to know when that happened and how many Ah's you drew out in time + amps. For example if the battery hit 10.5 volts at 10 hours then your battery capacity is now 50 Ah's or 10 hours X 5 amps = 50 Ah's, not 100 Ah's..
__________________
Marine How To Articles
Maine Sail is offline   Reply With Quote
Old 29-07-2011, 11:27   #7
Registered User
 
SV Demeter's Avatar

Join Date: Dec 2009
Location: Cruising Eastern Caribbean
Boat: Wauquiez Amphitrite 43- Demeter
Posts: 1,164
Re: Xantrex Link Pro Overestimating SOC Percentage

Great post Maine Sail as always thanks. Any suggestions for making a variable draw device to use for capacity testing? Maybe a halogen cabin light(s) with some sort of dimmer switch? Would a household 120v type dimmer switch work? Also wondering about measuring voltage during the capacity test. If I have this light set up to draw say 5 amps and I have my multimeter on the battery being tested. Wont the voltage readings be affected by the load from the light? ie wont this not be a "resting voltage"? Or does that not matter in this exercise? Lastly if I have a bank of (2) six volt wet cells and want to do a capacity test on each battery dont I need a 6 volt draw? Will my halogen cabin light and dimmer switch work here?
__________________
SV Demeter is offline   Reply With Quote
Old 29-07-2011, 11:46   #8
Senior Cruiser

Cruisers Forum Supporter

Join Date: Mar 2007
Location: Santa Cruz
Boat: Boatless Again
Posts: 4,310
Re: Xantrex Link Pro Overestimating SOC Percentage

Most 120v dimmer switches are SCR variety which clip off the AC waveforms to dim--they will not work with DC. Best to use DC lightbulbs. I haven't done the test in a while, but as best as I remember 10.5 volts endpoint is under the C20 load. If you have 6 volt batteries, you will have to use a 6V load if you want to test them separately--first do a 12v test, and measure the voltage across each battery at the end of the test--it should be roughly the same.
__________________
donradcliffe is online now   Reply With Quote
Old 29-07-2011, 11:49   #9
Registered User
 
SV Demeter's Avatar

Join Date: Dec 2009
Location: Cruising Eastern Caribbean
Boat: Wauquiez Amphitrite 43- Demeter
Posts: 1,164
Re: Xantrex Link Pro Overestimating SOC Percentage

Quote:
Originally Posted by donradcliffe View Post
Most 120v dimmer switches are SCR variety which clip off the AC waveforms to dim--they will not work with DC. Best to use DC lightbulbs. I haven't done the test in a while, but as best as I remember 10.5 volts endpoint is under the C20 load. If you have 6 volt batteries, you will have to use a 6V load if you want to test them separately--first do a 12v test, and measure the voltage across each battery at the end of the test--it should be roughly the same.

I was planning to use a 12v light, old halogen cabin light acutally. Was more curious about finding a 12v dimmer switch. Maybe I should not have given away those sensibulb dimmers I had while back. Where would i find a 6v load anyway?
__________________

__________________
SV Demeter is offline   Reply With Quote
Reply

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Xantrex Link Pro Sailagain Electrical: Batteries, Generators & Solar 52 19-11-2011 18:54



Copyright 2002- Social Knowledge, LLC All Rights Reserved.

All times are GMT -7. The time now is 10:23.


Google+
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Social Knowledge Networks
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.

ShowCase vBulletin Plugins by Drive Thru Online, Inc.