Need Help Setting up my Battery Monitor

[Jbaffoh]

Yes, the LinkPro has adjustable Peukerts and charge efficiency settings. The LinkPro's default Peukert just happens to be what Trojan recommends: 1.25. So far, I've left the charge efficiency set to auto.

As far as the "Gotcha," my float voltage setting (on the meter) is 3.6, which is the rated voltage of my solar controller (in reality, the meter reads 13.52 when the controller trickles down to .1 amp). The charge current is set to .5% (260 ah bank, solar controller max's at 3.5 amps). The time remains at the 240 second default.

My problem is not being fooled that the bank is fully charged. I'm being fooled into believing the bank is losing 1.5 ah per night, when it's fully charged.

[Stu Jackson]

Originally Posted by noelex 77

Battery return Amps is usuful way of a battery monitor resetting itself. IMHO it is a feature that should not be disabled, but setting of 13.2v is much too low and will result in silly readings.
Try setting the voltage to 0.2v below the bulk charge voltage value, the return amps to 1.5-2.0% and the time to a few minutes. I think you will find it a very useful and valuable feature of any battery monitor when more sensible parameters are used.
Dropping back the charging to a float value is also sensible when these parameters have been met. This is more critical with gel and AGM batteries, particularly when charged by solar, where the charge voltage can sometimes be just below the absorption threshold for long periods.
The return amps under float conditions will give you an idea if the parameters were correct.
The only slight problem with this is the battery monitor will read a couple of percent high, but this is preferable to the monitor getting a long way out of sinc.

Perhaps you have missed the entire issue discussed in the Gotcha topic, which is PREMATURE declaration of a full bank. Please remember that the algorithm counts KWhr, and converts it to amp hours.

In any event, the issue is not whether the monitor will show that the banks aren't full, but rather that WITHOUT CHANGING THE DEFAULT parameters of charging voltage and % capacity, the monitors could resynch by themselves and show the banks full when they are NOT.

That's all.

And the reasoning behind it is described, with simple math, in that topic.

So, the suggestion is to modify the two parameters to avoid premature declaration of a full bank.

The resynch can be done manually anytime the skipper knows the bank is full.

Maine Sail, Rich, Donalex and I have done quite a bit of research on this and that's why we've made these recommendations.

In the ybw.com forum from the UK, we had a lot of discussion on this topic, and some folks simply couldn't believe you would have to change parameters. What we had to explain was the math discussed in the topic. Some of them still couldn't get it.

Many, many skippers have nominal 400 ah banks.

From that topic:

Re: Link-series Charging Algorithms
« Reply #3 on: April 23, 2009, 09:32:13 AM »

No, Ken, you're not missing anything. It WILL eventually register full because the way it works is that it does count the AH down by using the kWHr function internally. The "trick" here is ONLY to prevent it showing full prematurely with the "default" 13.2 V and 2% "calculators," and not never ever showing "full." It will show full when the kWHrs counts to zero, and WON'T do it PREMATURELY with the float voltage and percentage of charge "method."

The Link algorithm has TWO ways of showing full: 1) the 13.2V and 2% method; and 2) counting the amp hours (really kWhrs) down to zero. It's the first of these that can make it say FULL prematurely. The idea is to avoid the first and ONLY use the second, which is when the algorithm actually counts the input amperage back to zero, meaning "I'm really full!"

What we need to do is to think of it in two different ways:

1. Your concern: "My Link will never show it is full."
vs.
2. Rich's concern: That the Link will show it IS full BEFORE IT REALLY IS, and therefore, if operated either manually or in my case automatically, will undercharge the banks.

These are two very important and different things. It's kinda like thinking of going up stairs or down stairs.

Rich wrote: "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."

He's NOT saying it never will say full, he's saying that it will NOT show a "full" display PREMATURELY based on the recognition of the 2% and 132.V"matching" for 4 minutes (regardless of what the amp hour/kWhr counter is doing) and will not say full until the ah are counted down to zero.

This means there are two ways for the Link to say full: 1) the 13.2V and 2% of ah capacity over 4 minutes (which we've shown can mathematically happen when your fridge kicks on after a daysail) ; and 2) when the ah counter gets to zero. Item 1) is the "premature" real world possibility unless you make the changes suggested here. Item 2) is what you WANT to actually happen all the time.

What's happening inside the Link it that it is using kWHr to do the actual measurements and converting them to AHs for the display.

The Link WILL eventually count the AH down to zero, but simply will NOT BE FOOLED into prematurely declaring FULL too soon (manually) or tripping into float (when connected to a Freedom inverter/charger) when the example Rich used, which I have seen myself, happens. It will ONLY use the amp hour countdown as the method of saying FULL. And that's all you need or want it to do.

This is when "one of the refrigeration units turns on drawing 5A" when the charger is in float and the current going into the bank is 2% of your 420 AH bank, or 8.4 amps. This WILL occur, say, when you've done a ONLY a four or six hour out and back to the dock daysail.

Condition: Your bank is only down 20 AH and you pull back into your slip and keep your fridge on and plug back in. Because of the battery acceptance (see that thread again) only a small amount of current, regardless of how big your charger is, is ever going to be "allowed" to flow back into the bank. My experience is that the charging current could easily and quickly reach that low of 8.4 amps (your case, or in my case 7.2 A = 360 X 2%).

What happens in real life is that the charger almost immediately kicks from bulk to absorption right into the float mode of 13.2 or 13.4 V because of the low acceptance of the batteries. And then the fridge kicks in drawing 5 A. Even without your Link, take a full bank after being charged overnight. You could drain your bank by running the fridge for an hour, and the first hour the fridge runs almost continually, so let's say it takes out that 20 ah. Turn on the charger. It'll go B-A-and right into float almost immediately. That happens all the time when the banks are within that last 15% of being charged.

At the time this occurs and after the next 4 or 5 or 6 minutes (the "specified" time period) of this condition, the Link says it is full. That's because the default 13.2 or 13.4 V is reached (you're in float mode right away) AND the 2% charge is being met because of low current acceptance exacerbated by the fridge making the current reading even LESS than the % (8.4 A less 5 A).

That's because the Link is always measuring the NET current flow, not only what the charger is producing. Flow from the charger (8 A) LESS the instantaneous current draw of the fridge (5A) = 3 amps, which is what the Link is seeing. Turn on some more lights or other DC appliances and it'll be even less, right?

Again, based on that acceptance thread, the batteries are NOT full. You've only been plugged in for 10 minutes, and your fridge is running, but the Link says full - NOT! This is a very real situation.

Some people also mistakenly think that a charger will increase its amperage when the fridge kicks in. This is NOT the case, because chargers simply aren't that smart - they produce a rising voltage until the set point, and the battery acceptance limits and reduces the current as the bank gets full.

The "math" can also happen at the end of every charge cycle, even if you start with a 50% depleted bank and you're running your fridge. For me, it'd kick the [Freedom combined inverter/charger] charger off PREMATURELY because the Link says the bank is full. That was the genesis of Rich's discussion, regardless of whether it was his generator with which he tried to minimize run-time, or someone being plugged in all the time with the Link 2000 running a Freedom inverter charger. It would also apply to anyone who is manually switching their charger off when the Link says it's full without making these changes to the default values.

If allowed to occur on a regular basis, those few "missing" ah start to add up in real life and real AHs, and chronically undercharged banks.

Rich is not saying it will never show full, he's saying that the concept is to avoid showing "full" prematurely when they are not, because the way the "guts" of the Link works, it WILL eventually and correctly show full when the ah are counted down to zero. This is based on the kWHr calculation within the algorithm. By simply changing the full parameters will assure that the bank is actually full, and not before then.

The "synchronization" issue you mentioned is pretty much the same for both your Link Lite and our Link 2000, and is not an issue in this concept. It is essentially a reset to zero concept.
****************************
****************************

While you could continue to operate your monitor as you describe, simply knowing how it works is very important.

Your boat, your choice.

The choice is simple: either modify the default parameters (which are MADE to be changed, BTW) OR don't and have the mnonitor resynch automatically but prematurely. If you're running a Freedom I/C off a Link 2000, you're gonna go into float prematurely, every time. Not so good.

Also, IIRC, you can't modify the time function.

Good luck.

[sailinglegend]

Originally Posted by Jbaffoh

Is this an ad?............ Or just marketing hype?

Sorry, but I have had problems over the years with the accuracy of my Battery Monitor so I have ordered a SmartGuage.

You were trying to account for an accuracy of 1.5 Ah. My point really was that you are wasting your time as battery monitors are so inaccurate. Yes you can set all the variables, but they change as the batteries age, The SmartGuage "Shouldn't be accurate" if it only measures battery volts, but the UK military seem to be happier with theirs than they were with Battery Monitors.

[noelex 77]

Originally Posted by Stu Jackson

Perhaps you have missed the entire issue discussed in the Gotcha topic, which is PREMATURE declaration of a full bank. *
.

I do unstandardised the issue and agree that the default values are poorly chosen, as I stated.
The link you posted suggests setting the parameters to *15 v and 1%, this completely *disables the battery return amp reset.*
I think this is also wrong. I proposed some different parameters, give them a try.

[goboatingnow]

I've been led to believe the LinkPro and similar monitors also use sophisticated algorithms based on precise measurements of millivolts across the shunt. What does the Smart Meter do differently?

However, while your advice is interesting, I've obviously purchased and installed my meter with it's shunt. My question was about finding realistic self-discharge rates and temperature coefficients for my Trojan T-145 batteries so I can improve the accuracy of my meter. Got advice? Or just marketing hype?

Smart Meter has no shunt for a start.

What smart Meter do is they model the very complex model of voltage versus state of charge that pertains to a lead acid battery. the accepted advice is that of course that measured voltage is not a good measure of SOC. In fact the case is that once you apply the model it is a very good indication of SOC. The computation is more complex and typically hasn't been attempted in battery monitors as the shunt system is simple computationally and easy to implement.

Dave

[Stu Jackson]

Originally Posted by noelex 77

I do unstandardised the issue and agree that the default values are poorly chosen, as I stated.
The link you posted suggests setting the parameters to *15 v and 1%, this completely *disables the battery return amp reset.*
I think this is also wrong. I proposed some different parameters, give them a try.

Sorry, but that's simply not true. What changing those default parameters does is simply stop the meter from declaring the bank full PREMATURELY.

It does NOT stop the KWhr meter from counting to zero, which is what you DO want to happen.

[noelex 77]

Originally Posted by Stu Jackson

Sorry, but that's simply not true. What changing those default parameters does is simply stop the meter from declaring the bank full PREMATURELY.

It does NOT stop the KWhr meter from counting to zero, which is what you DO want to happen.

I think if you reread the post you will find it is accurate, note I said it disables the battery return amps reset.

A good battery monitor will and should countdown to zero using both battery return amps and total energy (AHrs or KwHrs).
Setting the parameters you suggest ( 15v 1%) completely disables one of the mechanisms( battery return amps), because the parameters will never be met.


The default values for the battery return amps are totally wrong.This is why it does not work and declares the bank full prematurely. If you set these to appropriate values both methods will work correctly. This is much better than disabling the feature completely.
With both methods working correctly the meter will be more accurate.

[sailinglegend]

Originally Posted by noelex 77

.....Setting the parameters you suggest ( 15v 1%) completely disables one of the mechanisms( battery return amps), because the parameters will never be met......

If you do this then for very many boats, where the batteries are NEVER fully charged, the battery monitor will never show any useful or near useful information. Talk to the BM manufacturers and they will tell you that is why they set values for Ah reset that they know are too low - but that way the monitors will be half useful.

[Stu Jackson]

WADR to the last two posts, you are both missing the point. Half of the picture is not worth anything.

The issue is that the monitor WILL resynch to zero when the batteries are full and not before.

I apologize that the linked topic is so long, but it does cover your suppositions, which are simply incorrect.

Would you like to have a fuel gauge on your car say the tank is empty before it really is?

Defaults are just that, and were chosen by the software engineers as a simple starting point, NOT necessarily because they were correct for every conceivable condition or battery capacity or float voltage.

Again, if you have a Freedom I/C controlled by a Link 2000, the charger WILL go into float prematurely, significantly extending the charging time for your bank.

All we're suggesting is that you understand how the algorithm works and modify the defaults to avoid premature declaration of full.

A halfway meter is less than useless. "Specially if you know how to adjust it properly. Why ignore that built-in feature, called modifying defaults?

The Links WILL declare full batteries when the KWhrs do count down to zero, which is the ONLY thing that should be an indicator of full batteries.

[noelex 77]

Originally Posted by Stu Jackson

The Links WILL declare full batteries when the KWhrs do count down to zero, which is the ONLY thing that should be an indicator of full batteries.

The problem with this is that over time the amp hour count will gradually drift out of sync. The better monitors include battery return amps to reset this drift. When the voltage, return amps, and time are correctly set the system works perfectly. It is also the most accurate method of terminating the absorption charge phase, particularly with a solar system where the battery voltages *can sometimes be just under the absorption threshold for long periods.

Don't let the poor choice of the Links default values put you off. Try it with some sensible values instead of the defaults (0.2 v below bulk) and you find it works well.*

Originally Posted by Stu Jackson

Would you like to have a fuel gauge on your car say the tank is empty before it really is?

To use this analogy. Imagine the fuel gage in your car does not read the fuel level, but measures the fuel used by the engine and subtracts the fuel that is put into the tank. This is essentially how the battery monitor works.
It would be sensible to also have a sensor that registers just before the fuel tank overflows and use this to reset the gauge to full. This is how return amps work.

[Stu Jackson]

Not sure I understand. Our battery monitor works on "return amps" too, since it reads NET input and output.

Seems the only conceptual difference we have is that I prefer to reset my monitor to zero manually when I know the bank is full (i.e., + amp hours showing) where you prefer to modify the default parameters less than we've chosen to do.

You wrote: Don't let the poor choice of the Links default values put you off.

I don't at all. I don't think they're a poor choice at all. They are simply a default, mean to be modified based on your (house) bank size and the gadgets you have on your boat. Heck, if we didn't have a fridge, and had a 200 ah house bank, and no Freedom 15 I/C, there wouldn't be a need to modify anything. As I suggested earlier and in the algorithm topic, you simply have to "do the math" to find out what works for you.

Years ago, when programming first became available on mainframe computers (heck, 'member those days?!?) the program used to calculate the air quantities didn't make sense to me, 'cuz they were wrong when i did it by hand. When I talked to the programmer he said: "Oh, those are just the default values." I said, "Sure and they only work in one city in the entire USA, what happens when you are in a different city?" "Oh, you change the parameters." "Right, fella, but where have you bothered to tell anyone about this little gem?!?"

Defaults, the curse of unknowing engineers and sailors!

The issue remains, of course, declaration of prematurely full.

Your boat, your monitor, your choice, but I gotta tell ya that 99% of people installing battery monitors, professionally or personally, don't have a clue about these features.

Thanks for listening.

Stu

[noelex 77]

Originally Posted by Stu Jackson

Not sure I understand. Our battery monitor works on "return amps" too, since it reads NET input and output.

Return amps and net in and out are two very different things.
Return amps say if the voltage is high for a reasonable time when the current going into the batteries is low ( the return amps ) the batteries are full or close to full.
Net amp hours in and out ( or KWHrs for some ) counts the power in and out and says the batteries are full when the net result is zero.

Good battery monitors use both systems. By setting the parameters at 15v ( which will never be reached) you have deliberately disabled the return amp reset completely, leaving only the amp hour ( or KwHr in this case) countdown to reset the monitor.
I understand why you did this, because the Link default values mean that the return amp setting was reseting the counter prematurely. This is not surprising the Link defaults @ 13.2v and 2% is way too low. If you try some more sensible values for the return amp setting ( 0.2v less than the bulk charge rate and around 1-2% as a starting point) the return amp reset feature will be retained, but will work properly.
The battery monitor will not need to be manually reset often, or at all, and more importantly the charge termination from absorption to float will be more accurate. ( with a charger, or controller that is in communication with the monitor.)
This advice does not just apply to Link monitors. Other monitors , battery chargers and solar controllers use return amps, it is an expensive feature that is used in better equipment. Adjust it correctly and it will work well.

[Stu Jackson]

Thanks for that great explanation, now I understand the difference.

1. If there is no draw on the system, would return amps be the same for the Link with a single or dual shunt (the Link 2000 of course has a dual shunt, but the second one as you know is just for the reserve bank), so just deal with a single shunt concept.

2. Do monitors with return sensing have separate shunts? How does this work, I'm still not clear on the differences.

3. What monitors do this? Or are you saying that by changing the parameters LESS than I have suggested, the Link will do so?

It's early for me here, so I'll also think some more on it.

Thanks,

Stu

[noelex 77]

Originally Posted by Stu Jackson

Thanks for that great explanation, now I understand the difference.

1. If there is no draw on the system, would return amps be the same for the Link with a single or dual shunt (the Link 2000 of course has a dual shunt, but the second one as you know is just for the reserve bank), so just deal with a single shunt concept.

2. Do monitors with return sensing have separate shunts? How does this work, I'm still not clear on the differences.

3. What monitors do this? Or are you saying that by changing the parameters LESS than I have suggested, the Link will do so?

It's early for me here, so I'll also think some more on it.

Thanks,

Stu

Battery monitors only need a single shunt to measure the return amps. The return amps are simply the amps returning to the battery.
The complication occurs with battery chargers and solar controllers. These use a shunt to measure the charging output, but this single shunt gives no information how much of this current is going into the battery. Some of the current will be used on the load.
So a solar regulator that is putting out 10A and measures a battery voltage of 14.4 v has no way of knowing if most of the 10A is running the fridge and other loads ( in which case the battery is close to full ) or if most of the 10A is going directly into the batteries ( in which case the batteries need more charge and should stay on the absorption voltage)
This lack of information can be solved in two ways.

1. The charger or solar controller can have the provision for two shunts. The second shunt measures how much of the charge current is reaching the batteries.
2. The charger or controler can communicate with battery monitor, via an interface that allows the charge source to access the information from the battery monitors shunt.

Both these systems alow the charger or controler to terminate the absorption phase of charge based on the return amps.
If the battery voltage is high, the current into the battery is low, and this is stable for a few minutes, the battery must be at a high state of charge.
This information can be used to terminate the absorption phase of charge, or reset a battery monitor.

The problem with Link system is the default values generate a premature termination and reset of the battery monitor, but if the default values are changed to appropriate settings the return amps feature can be adjusted to operate correctly.

Systems like this generally involve some sort of integrated approach that is more suited to more complex boats especially with batteries that are sensitive to correct charging like Gel and AGM.. It also involves the necessity to set the parameters correctly.

Integrated systems like this are available from Outback, Midnite, and Link.
Other companies make individual opponents that will use return amps like Xantrex.
The above companies also make battery monitors that use return amps. Other companies hint that their monitors use return amps, without telling customers exactly how they work ( I have still not worked out how the very common ( in europe) NASA battery monitors work or example).

[Stu Jackson]

Thanks, that was very interesting and informative. It is enlightening to see that various manufacturers are now using different methods for measurements and calibration. I'd be very interested in what you learn about those other systems and any links you can provide.

Thanks again,

Stu