PDA

View Full Version : Electrical System



BobJ
03-03-2008, 10:51 AM
I've been working on this and would welcome some opinions.

My system (same as for the 2006 race) is two Group 27 AGM's (92 Ah each) monitored with a Link 20, two Kyocera solar panels (43 watts/2.48A each) going through a Flexcharge PV7D dual-bank controller, and the stock Hitachi 35A alternator on my Yanmar 1GM10 as my primary charging method (I also took a spare alternator, belts, etc.) In 2006 I only used about three gallons of diesel to charge the batteries (over 17 days) and am pretty certain I could have avoided running it at all.

The puzzling element in the system is the Link 20. To power itself, it draws from Battery #1. Over the past seven days, the Link shows #1's voltage dropped from 12.55 to 12.35 (with everything else turned off). The puzzling part is that it shows Battery #2's voltage remained constant at 12.7 but it shows #2 is down 11.2 Ah over the seven days. It shows Battery #1 was only down 1.9 Ah. This seems backwards. Similarly, #1 still shows four green bars (on the Link), but #2 is down to three yellow bars. In the 2006 race, the yellow bars on the Link 20 prompted me to run the engine when I probably didn't need to.

I've swapped batteries (#1 vs #2), returned one when it was still new (it tested full back at the store), and I've talked with the tech dept. at Xantrex (mfr of the Link). They said their monitors either work or they don't - they don't tend to give you bad data.

My current plan is to use my three-stage shore charger to fully recharge the two batteries, then run a full reset on the Link 20 and try it all yet again.

For now my questions are these:

1) Has anyone experienced similar inconsistent readings from a Link 20?

2) Is there a preferred contact point to place a voltmeter's probes to regularly measure the batteries' voltage, besides directly on the batteries? (It is possible but difficult to regularly access them underway.) I want to make sure I'm getting an accurate reading, independent of the Link 20.

3) Any comments on my electrical system overall?

I hope this isn't too confusing - I'm trying to describe something I don't understand very well.

Bob J.

tiger beetle
03-03-2008, 01:39 PM
2) Is there a preferred contact point to place a voltmeter's probes to regularly measure the batteries' voltage, besides directly on the batteries? (It is possible but difficult to regularly access them underway.) I want to make sure I'm getting an accurate reading, independent of the Link 20.

3) Any comments on my electrical system overall?

I can't comment on the Link 20 as I don't own one. I do have an Ample Power energy monitor, which provides a plethora of data and I only use two pieces: battery voltage (I check often), and ampere flow rate (I check occasionally). All other data regarding 'state' of battery charge I disregard. In essence I'm using the energy monitor as a very accurate, calibrated, temperature-compensated voltage and ampere meter (which it is).

My first suggestion is to get an accurate voltmeter, I recommend the Fluke meters, and use it to check voltage at the instrument panel (not the battery). E.g., you could hardwire the voltmeter to the panel-end of the battery cables and know the voltage your instruments have to work with - this will tell you when voltage has dropped too low to be useful (e.g., start to charge at 11.9 or 12.0 volts).

For accurate charge control purposes you really want to be measuring voltage and temperature at the battery terminal, which is most likely what the Link20 should be doing. You can read the voltage at the panel-end of the battery cables on a Fluke and assuming you do not see a significant drop in voltage across the cable you'll have a reasonable idea of the state of battery charge. What you really don't want to do is fry the AGM batteries.

Does this help at all?

- rob

BobJ
03-04-2008, 10:11 AM
Yes Rob, that's helpful. I have a good digital meter (not a Fluke though) and yes, the wires on the input side of the Link's shunt come directly from the battery terminals. Regarding measuring at the panel, I doubt I'm losing anything through the cables since they are oversized Ancor cables and relatively new, but I will check that.

I also found a Troubleshooting page on the Xantrex website. (It's amazing what you find when you Google "Link 20"!) It sounds like instead of charging the batteries up and resetting the Link, I need to draw the bank down by at least 10% and then recharge it. That will cause the Link to "recalculate charge efficiency" and stop "accumulating negative amp hours."

tiger beetle
03-04-2008, 12:43 PM
I also found a Troubleshooting page on the Xantrex website. (It's amazing what you find when you Google "Link 20"!) It sounds like instead of charging the batteries up and resetting the Link, I need to draw the bank down by at least 10% and then recharge it. That will cause the Link to "recalculate charge efficiency" and stop "accumulating negative amp hours."

That's the part of the Ample Power energy monitor that I ignore; it doesn't matter to me at all what the little unit has calculated as regards amp-hours remaining, amp-hours consumed, charge efficiency, peukert's equation, etc... What does matter is a) when exctracting power from the battery the voltage the battery can supply is key, and b) when charging the voltage and amperage of the charging source is critical.

What the Link 20 thinks about amps consumed, used, charge efficiency, negative, positive, or sideways is of no relevance when it comes to making the battery do work.

- rob