My third SmartGauge...

[Sea Dog]

3 minutes ago, Peter Thornton said:

I don’t think the fitting and operation is a “faff”, it’s just that it looks as if there were some faulty units around. Mine reads exactly the same as a good quality meter I have and reads 12.8 on a fully charged bank, which suggests that all is well.

I'm not sure it was as easy as that when MtB was having his problems Peter, but I forget the detail now as the debate went on for so long.  If the Smartguage really was that simple, how come it generates so much discussion?  I think you're about right - the less you know, the better an aid it is. This is no bad thing imho.

 

I assess my battery state primarily by voltage when discharging and tail current when charging, but that's not for everyone and I certainly don't trust my BMV SOC reading, although the Amp Hours Used reading may be a little more use.  For a simple SOC readout I guess the SG is hard to beat so far.

[Tony Brooks]

48 minutes ago, Peter Thornton said:

Our boat is a share boat and our owners fall into two camps: 

 

1. Owners who are intensely interested in batteries and spend a lot of time looking at numbers and voltages. 

2. Owners who would rather look at the view, don’t want to think about batteries on their holidays and just want to know when the engine needs running or they need to turn some lights off.

 

I’ve come to the conclusion that the Smartgauge is a great aid for the second group. 

 

It’s the first group (which includes me) who ask all the complicated questions and who would actually probably be better served by an Amp Hour counter.

 

That’s more or less the conclusion I came to, but I wondered if anyone had evidence other than reading the manual?

 

I think not, at least not a straight Amp hour counter. But, yes, one that also reads amps, volts and Ah out will give more information to those who can be bothered to study a bit but so far, apart from the Datacell that is far to expensive for most of us, any Ah left or charge percentage on such gauges is best ignored as likely to be misleading and battery destroying.

[MtB]

47 minutes ago, Sea Dog said:

Well, checking the calibration is a faff and how can you trust it unless you know is right?  Pondering where you are with it because no-one really knows how it works is a nuisance, if not a faff. MtB has 3 he doesn't trust.

 

It's not a case of me not trusting mine. Having wrecked a set of batteries by trusting my first, I checked its calibration and found it wrong. Once properly calibrated it works fine. Same happened on my second one. My third was correct from the factory. I trust all of them now I have established their true state of calibration.

 

It is the case that buyers of Smartgauges would be prudent not to trust theirs without checking their calibration first. Chances are now that a new one will be fine, but they won't know for sure unless they check. 

 

 

 

 

Edited October 29, 2018 by Mike the Boilerman
Add a bit.

[WotEver]

7 minutes ago, Mike the Boilerman said:

It is the case that buyers of Smartgauges would be prudent not to trust theirs without checking their calibration first. Chances are now that a new one will be fine, but they won't know for sure unless they check. 

True. Sad, but true. 

37 minutes ago, Sea Dog said:

If the Smartguage really was that simple, how come it generates so much discussion?

Two reasons. The first is that many folk (exclusively those who have never owned one) doubt that it can work. The second is that it shouldn’t be relied upon to determine 100% SoC when charging, which although it’s in the manual isn’t very prominent. 

[MtB]

Oh yes, and not to trust ANY smartgauge when it says the batts are 100% charged. The display may or may not be correct on this point.

 

Mine says 100% charged when the tail current gets down to about 5%. 

 

 

[Old Son]

My boat is on a mooring connected to shore power. The SG shows 100% charged. I also have a digital amp meter that was showing 0.0 which would be right because everything was turned off and the batteries fully charged. This last weekend I turned off the shore power and about 90 minutes later started the engine to warm it up before changing the oil. I checked the SG and the amp meter, both had not moved so there was no power from the alternator going into the bank. I was pleased with the results, it shows everything is quite accurate.

 

I run 24V leisure and 12V starter so am not able to check the state of the starter battery from my gauges.

[Sea Dog]

5 hours ago, Mike the Boilerman said:

It's not a case of me not trusting mine. 

No, I agree, but you're a good example as no-one can say, having bought 3 of them, you didn't give the thing a chance!

 

[Dr Bob]

6 hours ago, WotEver said:

 

Two reasons. The first is that many folk (exclusively those who have never owned one) doubt that it can work. The second is that it shouldn’t be relied upon to determine 100% SoC when charging, which although it’s in the manual isn’t very prominent. 

I think then that that is one reason. I am in the 'never owned ' category and doubt it can work ie shouldnt be relied upon to determine 100% SoC when charging based on my experience of developing global models based on linear regression or partial least squares techniques. I can quite understand it works well on discharge as I can see the relationship between voltage under load and SoC and hence a global model should be possible. On charge though, when nearing full charge, the voltage relationship is going to be very difficult to predict exactly across all the different senarios of battery type, make and charging devices and clearly the global SG model does not predict in this region well. My doubts are therefore well founded. The debate  from my perspective is then about how useful a device is that may not be accurate if looking to see when to stop charging. The only way to be certain to make that decision is via tail current hence for the knowledgable, spending £100+ on a meter is best spent on a battery monitor. I agree with you that for the less well informed, an SG is a useful tool if you follow the advice in the manual.

[cuthound]

55 minutes ago, Dr Bob said:

I think then that that is one reason. I am in the 'never owned ' category and doubt it can work ie shouldnt be relied upon to determine 100% SoC when charging based on my experience of developing global models based on linear regression or partial least squares techniques. I can quite understand it works well on discharge as I can see the relationship between voltage under load and SoC and hence a global model should be possible. On charge though, when nearing full charge, the voltage relationship is going to be very difficult to predict exactly across all the different senarios of battery type, make and charging devices and clearly the global SG model does not predict in this region well. My doubts are therefore well founded. The debate  from my perspective is then about how useful a device is that may not be accurate if looking to see when to stop charging. The only way to be certain to make that decision is via tail current hence for the knowledgable, spending £100+ on a meter is best spent on a battery monitor. I agree with you that for the less well informed, an SG is a useful tool if you follow the advice in the manual.

 

Continuing to charge for 4 hours after the SG has first indicated 100% should ensure a fully charged battery and does not involve any capital outlay, but would entail excessive expenditure of diesel.

[dmr]

1 hour ago, Dr Bob said:

I think then that that is one reason. I am in the 'never owned ' category and doubt it can work ie shouldnt be relied upon to determine 100% SoC when charging based on my experience of developing global models based on linear regression or partial least squares techniques. I can quite understand it works well on discharge as I can see the relationship between voltage under load and SoC and hence a global model should be possible. On charge though, when nearing full charge, the voltage relationship is going to be very difficult to predict exactly across all the different senarios of battery type, make and charging devices and clearly the global SG model does not predict in this region well. My doubts are therefore well founded. The debate  from my perspective is then about how useful a device is that may not be accurate if looking to see when to stop charging. The only way to be certain to make that decision is via tail current hence for the knowledgable, spending £100+ on a meter is best spent on a battery monitor. I agree with you that for the less well informed, an SG is a useful tool if you follow the advice in the manual.

Not sure why we need to understand linear regression to comment on the Smartgage, but here goes....why stop at linear?, we could use a quadratic, or even a cubic, in fact its possible to derive the equations for a least squares fit to any arbitrary function.....if anybody want to know how just let me know ?.

 

The other qualification required  to criticise the Smartgage on this form is "Not Actually Owning One". I fail here as I've owned one for years. I was a non believer for a long while but a couple of posters on this forum kept telling me to stop getting into a mess with batteries and to buy a Smartgage.... I think they were called SmileyPete and Gibbo  ?

The only relevant fact is....the Smartgage works!, its tell you how much 'leccy you have left in your batteries with adequate accuracy. What more is there to worry about?....but if you don't want one then don't buy one, though personally I can't see how a liveaboard boater can do without one. It has a few limitations, its not so accurate when charging, and very inaccurate at indicating full charge but was not designed to do this, though I do still find it quite useful during charging, especially if doing a routine 50% to 80% type charge.  I really can't be doing with turning everything off and measuring rested voltage    ?

 

.....most looked at instruments on my boat

1 Smartgage

2 Charge Current

3 Discharge Current

3 Engine Temperature

4 Oil pressure

5 Charge Voltage

 

..................Dave

[rusty69]

2 minutes ago, dmr said:

I was a non believer for a long while but a couple of posters on this forum kept telling me to stop getting into a mess with batteries and to buy a Smartgage.... I think they were called SmileyPete and Gibbo  ?

Well, to be fair, Gibbo would try and sell you one!

[dmr]

9 minutes ago, rusty69 said:

Well, to be fair, Gibbo would try and sell you one!

And for all the years I thought he was just giving impartial friendly advice ?

 

............Dave  (just checked, the Smartgage says 92%, definitely not fully charged yet)

[MtB]

31 minutes ago, dmr said:

And for all the years I thought he was just giving impartial friendly advice ?

 

............Dave  (just checked, the Smartgage says 92%, definitely not fully charged yet)

 

Dave, as you seem to be charging now, could you take a note of the tail current just as the SG clicks up to displaying 100% please? 

 

Many thanks!

[dmr]

27 minutes ago, Mike the Boilerman said:

 

Dave, as you seem to be charging now, could you take a note of the tail current just as the SG clicks up to displaying 100% please? 

 

Many thanks!

Bludy 'ell Mike that's a co-incidence, I've just done exactly that and now read your post.

It got to 100% with a tail current of 17 amps. My battery bank is a nominal 700Ah so that's 2.4% so its not too bad. The Adverc was on the high part of its cycle so 14.8 volts, if it was on its 14.4v phase then the charge current would be quite a bit lower.    I expect in reality the capacity is a bit less than 700Ah, and also the tail current will vary quite a bit depending upon where I am in the "sulphation cycle".

 

We haven't had a cruise for over a week now so I reckon I will have just a bit of sulphation. Serious boating tomoro though, we need to wind and that's going to involve doing a significant flight of locks down and then back up again.

 

...............Dave

[WotEver]

2 hours ago, dmr said:

The other qualification required  to criticise the Smartgage on this forum is "Not Actually Owning One".

Yup, as I noted earlier.

[WotEver]

2 hours ago, dmr said:

Not sure why we need to understand linear regression to comment on the Smartgage...

Besides, Gibbo informed me ages ago that Bob’s guess at how SmartGauge works is wrong anyway. 

[Peter Thornton]

25 minutes ago, WotEver said:

Besides, Gibbo informed me ages ago that Bob’s guess at how SmartGauge works is wrong anyway. 

Have to say, I don’t really understand the principle on which they work. We have been using a voltmeter to monitor SOC and, as anyone who’s tried this knows, the problem is that the volts measured varies with the load on the batteries. So I guess that one function of the Smartgauge is to smooth all of these variations out. But without knowing what load is on the batteries - i.e. the amps being drawn - how does it learn the characteristics of your battery bank and deduce the SOC?

One thing I’ve noticed, after having it for all of 30 hours, is that it stays at 100% for quite a long time after charging stops and current draw begins.

But I like its simplicity and think it might be very useful on our Share Boat.

Incidentally, I spent Saturday night at a marina social event and met a couple who had just embarked upon a Livaboard Life. They seemed a little suprised when I congratulated them on their new life and said that it would now be dominated by batteries and toilets .............

[WotEver]

19 minutes ago, Peter Thornton said:

They seemed a little suprised when I congratulated them on their new life and said that it would now be dominated by batteries and toilets .............

And pretty fans. Mustn’t forget the rattly fans...

[Dr Bob]

1 hour ago, WotEver said:

Besides, Gibbo informed me ages ago that Bob’s guess at how SmartGauge works is wrong anyway. 

Now I know you can't/won't say what the basis of the model is but can you give me a clue as to why my 'guess' is wrong? Genuinely interested as one of the hangovers of the modeling business I ran predicting MON and RON of gasoline and Cetane rating of gas oil is to understand better methods of modeling.

[WotEver]

18 minutes ago, Dr Bob said:

Now I know you can't/won't say what the basis of the model is but can you give me a clue as to why my 'guess' is wrong? Genuinely interested as one of the hangovers of the modeling business I ran predicting MON and RON of gasoline and Cetane rating of gas oil is to understand better methods of modeling.

Ummmm.... no. 

 

Not because I choose not to but because I don’t know myself. It just works and I’m happy with that. 

Edited October 29, 2018 by WotEver

[MtB]

42 minutes ago, WotEver said:

Ummmm.... no. 

 

Not because I choose not to but because I don’t know myself. It just works and I’m happy with that. 

 

Errrr.....

 

if if you don’t know the basis of the modelling how can you say Dr Bob’s guess is wrong? 

[Richard10002]

1 hour ago, Peter Thornton said:

One thing I’ve noticed, after having it for all of 30 hours, is that it stays at 100% for quite a long time after charging stops and current draw begins.

But I like its simplicity and think it might be very useful on our Share Boat.

Whatever you do, don’t base stopping charging on the fact that it reads 100%, and don’t trust it, when discharging, until it reads 80% or below.

 

The manual states/intimates the former, and somebody here said that Gibbo told them the latter. They also said that he had added something like, “why would anyone need to know the SoC when discharging, if it’s over 80%”

[dmr]

1 hour ago, Peter Thornton said:

Have to say, I don’t really understand the principle on which they work. We have been using a voltmeter to monitor SOC and, as anyone who’s tried this knows, the problem is that the volts measured varies with the load on the batteries. So I guess that one function of the Smartgauge is to smooth all of these variations out. But without knowing what load is on the batteries - i.e. the amps being drawn - how does it learn the characteristics of your battery bank and deduce the SOC?

One thing I’ve noticed, after having it for all of 30 hours, is that it stays at 100% for quite a long time after charging stops and current draw begins.

But I like its simplicity and think it might be very useful on our Share Boat.

Incidentally, I spent Saturday night at a marina social event and met a couple who had just embarked upon a Livaboard Life. They seemed a little suprised when I congratulated them on their new life and said that it would now be dominated by batteries and toilets .............

 

Yes, Smartgage does take quite a while to respond after charging stops, this is almost certainly due to surface charge.....the battery voltage is artificially high due to the charging, this is a basic physical effect so not much Smartgage can do about it. Those who like to do it themselves by turning everything off and measuring the "rested voltage" have exactly the same problem.

 

There is lots of stuff on this forum about how the SG might work. Best guess is that it knows the current flow and knows how to compensate for the effect this has on voltage. When something is turned on/off the change in current causes a change in voltage (in quite a complicated way) and SG accurately monitors these voltages changes to do a good guess at the current.

 

............Dave

[Dr Bob]

7 minutes ago, Mike the Boilerman said:

 

Errrr.....

 

if if you don’t know the basis of the modelling how can you say Dr Bob’s guess is wrong? 

I am not even going to try to argue!

[WotEver]

9 hours ago, Mike the Boilerman said:

 

Errrr.....

 

if if you don’t know the basis of the modelling how can you say Dr Bob’s guess is wrong? 

I didn’t say that. Gibbo said that.