All Electric?

[nicknorman]

9 hours ago, Peter Laursen said:

Having Googled my day away in search of advice on Empirbus CLC I find myself as new member of the forum on Canalworld.net. I hope, you will bear with me, even though my boat has a mast and normally navigates Danish waters - and not canals...

As far as I can tell, you have had the Empirbus CLC on your narrowboat for 10 + years? I just bought at boat (build 2009) with Empirebus CLC installed. One of the two membrane panels is failing. At a quoted price of EUR 1.000 (!) and no means of bypassing generally upon systems failure, as far as I can tell, I'm considering removing all of it and rewire the boat, if age is about to become a problem for the system. EUR 1000 would take me a fair bit 🙂. 

But I was wondering, what your experience with 10+ years of ownership is - and am I being to dramatic?
Kind regards Peter

Hi Peter

 

In the first year, one of the modules had a problem but it was replaced under warranty. Since then, the system has worked perfectly and when I have made various changes to the wiring, added new features etc, it has been very easy because I just have to change the programming.
As an example, we had a sump box with float switch for the shower drain. Really bad idea because after a few years the float switch stopped working and the sump box overflowed into the bilges. Yuk! So I installed a Whale Gulper, and I installed flow switches in the shower feed connected to an Empirbus node spare input. Changed the configuration so the pump output is triggered by the flow switches and has a delayed off so it keeps running for about 10 seconds after the shower is turned off. All works perfectly and no having to remember to turn a pump switch on and off.

 

I did get a spare membrane panel and a spare node with a couple of boards (input and output) when we bought the system, but a few years ago I added the spare  membrane panel into the system to switch some new functions. Not now having a spare worried me a bit but I was able to buy a second hand membrane panel from ebay for around £100 so once again I have a spare - just for peace of mind, since the panels are both still working fine.

 

You say your panel is failing, in what way is it failing?

[Peter Laursen]

1 hour ago, nicknorman said:

 

 

You say your panel is failing, in what way is it failing?

Hi Nick,

Thank you for taking time to answer.

Good to hear, you haven't encountered further problems. 

Here is a link to a video showing the panel "acting with a mind of it's own".....

https://www.dropbox.com/s/wvd375rkcb3tyzv/20220820_200334.mp4?dl=0

 

EmpirBus support in Sweden says, it's classic sign of a panel in need of replacement.

 

On a non professional level I know my way around electric installations etc. But I don't know enough about electronics to try and fix an error like this. If I knew, I only needed to replace a capacitor or the like -- no problem. But which one ha ha 🙂

 

[Tony Brooks]

Not directly on topic but it illustrates nicely how when all this electronic bus stuff goes wrong it is likely to cost ordinary folk a lot of money. It is fine for Nick who could probably design and make a good number of components. Give me wires and switches any day, much easier to diagnose and repair with readily available bits, even if it won't change your lighting colour or do fancy things with shower pumps that a pneumatic switch would do. Goodness knows what will happen when one of these bus linked lithium charging system plays up and how much it will cost to diagnose and put right. Horses for courses.

[Peter Laursen]

1 minute ago, Tony Brooks said:

Not directly on topic but it illustrates nicely how when all this electronic bus stuff goes wrong

Hi Tony,
Although I can certainly se the pros of a digitally switched system I wonder if I should consider "going old school". But that is also a rather big project on a boat, where the EmpirBUS is already installed. But good to hear, that Nick hasn't encountered more hardware errors.....

[Tony Brooks]

7 minutes ago, Peter Laursen said:

Hi Tony,
Although I can certainly se the pros of a digitally switched system I wonder if I should consider "going old school". But that is also a rather big project on a boat, where the EmpirBUS is already installed. But good to hear, that Nick hasn't encountered more hardware errors.....

 

If you can sort it for a cost affordable to you, it is probably not worth a total reinstallation, it will cost far more than that 1000 euro unless you totally DIY and then you face a problem if routing the many more fairly heavy cables required.

[nicknorman]

13 minutes ago, Peter Laursen said:

Hi Nick,

Thank you for taking time to answer.

Good to hear, you haven't encountered further problems. 

Here is a link to a video showing the panel "acting with a mind of it's own".....

https://www.dropbox.com/s/wvd375rkcb3tyzv/20220820_200334.mp4?dl=0

 

EmpirBus support in Sweden says, it's classic sign of a panel in need of replacement.

 

On a non professional level I know my way around electric installations etc. But I don't know enough about electronics to try and fix an error like this. If I knew, I only needed to replace a capacitor or the like -- no problem. But which one ha ha 🙂

 

Yes it certainly does look wonky (that is the technical term!). I have never seen anything like that behaviour. If it were me I’d check the CANBUS wiring - on the back of the panel are 4 wires, 2 for power and 2 for CANBUS (high and low) the CANBUS is a daisy chain of the 2 wires (ie goes to all the nodes and panels) but at each end there needs to be a terminating resistor, so again I’d check the terminating resistors are present and correct.

 

I am not going to argue with Empirbus support because they are the experts, but personally I’d want to check what data was actually on the bus to see if the panel was the cause of the problem, or simply reacting to rogue data on the bus. In any case, if you are going to cough up for a new panel, it will have to be configured and for that you need the programming USB adaptor and the (free) software. As well as programming devices, that allows you to monitor bus activity with the CANBUS messages translated into human-readable form. And if you do get a new panel, do you have the original configuration file? Do you know where the programming port is? I think you can extract the config from the existing panel but only if it is working!

 

Final thoughts, although it is encapsulated in resin at the back, at the front you can slide the button legend in and out. I don’t know if your panel is ever exposed to water but, whilst it is supposed to be reasonably waterproof, could it have become damp under the  outer membrane? Just wondering if taking it home, removing the legend strip and letting it dry out for a few days might help? Oh and do the buttons all have the same tactile feel? Just wondering if a button could stuck down or something.

[nicknorman]

18 minutes ago, Tony Brooks said:

Not directly on topic but it illustrates nicely how when all this electronic bus stuff goes wrong it is likely to cost ordinary folk a lot of money… Horses for courses.

Talking of horses, they said exactly the same thing when folk proposed replacing horses with infernal combustion engines. The proposal was to take a self-mending self-replicating entity that ran on grass, and replace it with a metal thing that was more expensive, noisier, needed stuff to be drilled out of the ground to run, and when it broke it had to be fixed by someone - but hardly anyone on the cut had the expertise to do it. Crazy!

[MtB]

4 minutes ago, nicknorman said:

Talking of horses, they said exactly the same thing when folk proposed replacing horses with infernal combustion engines. The proposal was to take a self-mending self-replicating entity that ran on grass, and replace it with a metal thing that was more expensive, noisier, needed stuff to be drilled out of the ground to run, and when it broke it had to be fixed by someone - but hardly anyone on the cut had the expertise to do it. Crazy!

 

 

Quite. 

 

It's one thing to point out how a custom designed system like Nick's would cost a fortune to get fixed by some random marina somewhere but that won't be happening, as the designer is on hand in Nick's boat all the time to fix anything that goes wrong with it. 

 

Once long into the future all boats are controlled by Canbus/Empirebus, then yer average boatyard workshop spod WILL be able to fix them, cheaply and easily because s/he deals with them all day, every day. Assuming we still have canals at all.

 

 

[nicknorman]

13 minutes ago, MtB said:

 

 

Quite. 

 

It's one thing to point out how a custom designed system like Nick's would cost a fortune to get fixed by some random marina somewhere but that won't be happening, as the designer is on hand in Nick's boat all the time to fix anything that goes wrong with it. 

 

Once long into the future all boats are controlled by Canbus/Empirebus, then yer average boatyard workshop spod WILL be able to fix them, cheaply and easily because s/he deals with them all day, every day. Assuming we still have canals at all.

 

 

Just for the benefit of the OP, the “custom designed system” Mike is talking about, isn’t the Empirbus system which is just an off the shelf system with reasonable support in the UK (Atlantis Marine Power), it’s the lithium battery management system.

 

And yes of course you are right in the main - just look at cars which are mostly run by CANBUS these days. Your average garage mechanic has sufficient knowledge to fix things - they just plug a computer in, it tells them what is wrong and which bit to replace or hit with a hammer. Which is a lot simpler than grovelling around with greasy spanners!

Edited August 23, 2022 by nicknorman

[Tony Brooks]

9 minutes ago, nicknorman said:

they just plug a computer in, it tells them what is wrong and which bit to replace or hit with a hammer

 

I would question that. It tells them what it THINKS is wrong and bits get replaced because the computer tells them to, but which would have been fixed if the correct terminal connection had been cleaned. The garage mechanics also get to know the more common faults both by being there and talking to colleges in other garages. The average boater does not, they are on their own.

[nicknorman]

12 minutes ago, Tony Brooks said:

 

I would question that. It tells them what it THINKS is wrong and bits get replaced because the computer tells them to, but which would have been fixed if the correct terminal connection had been cleaned. The garage mechanics also get to know the more common faults both by being there and talking to colleges in other garages. The average boater does not, they are on their own.

Yes in truth there has to be a certain amount of intelligence involved. I recently had a problem with the xenon headlights in my Skoda, they were just pointing down at the ground. Connecting the computer to the lighting module just said “databus error” which wasn’t very helpful. Then I though about it and remembered there were position sensors on the suspension to keep the headlights pointing at the right angle, so I redirected the computer to the chassis module and sure enough, one of the sensors was duff. Replaced the sensor and everything was fine.

 

But you are slightly conflating two things. Yes the average boater will never understand digital electronic boat systems. But then again the average boater doesn’t understand the basics of how a Diesel engine works/how to fault find it and how to fix it. Mike’s point is that boat technicians will evolve to deal with this sort of “hi tec” thing and it is already happening. Ed Shiers for example.

[Peter Laursen]

2 hours ago, nicknorman said:

Yes it certainly does look wonky (that is the technical term!). I have never seen anything like that behaviour. If it were me I’d check the CANBUS wiring - on the back of the panel are 4 wires, 2 for power and 2 for CANBUS (high and low) the CANBUS is a daisy chain of the 2 wires (ie goes to all the nodes and panels) but at each end there needs to be a terminating resistor, so again I’d check the terminating resistors are present and correct.

 

I am not going to argue with Empirbus support because they are the experts, but personally I’d want to check what data was actually on the bus to see if the panel was the cause of the problem, or simply reacting to rogue data on the bus. In any case, if you are going to cough up for a new panel, it will have to be configured and for that you need the programming USB adaptor and the (free) software. As well as programming devices, that allows you to monitor bus activity with the CANBUS messages translated into human-readable form. And if you do get a new panel, do you have the original configuration file? Do you know where the programming port is? I think you can extract the config from the existing panel but only if it is working!

 

Final thoughts, although it is encapsulated in resin at the back, at the front you can slide the button legend in and out. I don’t know if your panel is ever exposed to water but, whilst it is supposed to be reasonably waterproof, could it have become damp under the  outer membrane? Just wondering if taking it home, removing the legend strip and letting it dry out for a few days might help? Oh and do the buttons all have the same tactile feel? Just wondering if a button could stuck down or something.

 

Wonky.... good word! Just been to the boat prior to reading your message. I disconnected and reconnected the four wires on both panels. No corrosion, and they were properly mounted. The boat is in general in pristine shape being kept inside during the winter and with a full boat cover on it during the summer in a fresh water lake (all this will change now in Scandinavia :-)). So I don't expect to find moisture related problems etc. I will check termination on the bus, and I will log on to the system and see what I can find. Maybe there is some kind of error log. I have the passwords from the boat yard to the configuration file.

 

The panel acted in a new way when testing today:

At some point AFTER turning the on/off key at the chart table to OFF (which is connected to port 1 on the control unit), a green light in one of the buttons on the panel kept glowing. Shortly after, the background illumination came on (still with the main on/off key in OFF position). The was a distinct humming sound from the panel which went away, when I managed to turn it of again. Strange.... I would expect the panel to get its power from the main unit. So maybe the main unit is part of the problem...

Thanks for your inputs so far - much appreciated!
 

[nicknorman]

23 minutes ago, Peter Laursen said:

 

Wonky.... good word! Just been to the boat prior to reading your message. I disconnected and reconnected the four wires on both panels. No corrosion, and they were properly mounted. The boat is in general in pristine shape being kept inside during the winter and with a full boat cover on it during the summer in a fresh water lake (all this will change now in Scandinavia :-)). So I don't expect to find moisture related problems etc. I will check termination on the bus, and I will log on to the system and see what I can find. Maybe there is some kind of error log. I have the passwords from the boat yard to the configuration file.

 

The panel acted in a new way when testing today:

At some point AFTER turning the on/off key at the chart table to OFF (which is connected to port 1 on the control unit), a green light in one of the buttons on the panel kept glowing. Shortly after, the background illumination came on (still with the main on/off key in OFF position). The was a distinct humming sound from the panel which went away, when I managed to turn it of again. Strange.... I would expect the panel to get its power from the main unit. So maybe the main unit is part of the problem...

Thanks for your inputs so far - much appreciated!
 

On your last point, the membrane panels are permanently powered if the Nodes are powered. Typically there is a "master switch" input to one of the nodes, and most or all of the services subscribe to it, so that when you turn off the master switch you aren't removing power from the system, you are just forcing everything off except those things you want to remain on such as the bilge pump float switch power. When you press a membrane button to turn something on, the signal is still sent, but nothing happens if/because the virtual "master switch" is off. We use a membrane button for the master switch but I guess in your case it has been configured to use an input from the key switch.

 

There is a buzzer function on the membrane keypad but I've never used it. I do find that when the backlight is on, there is a bit of a  buzzing/humming sound for some reason  You can also have beeps with key presses etc

 

I suppose another question is, if you disconnect that bad membrane panel, does everything else work exactly as expected?

Edited August 23, 2022 by nicknorman

[Peter Laursen]

6 minutes ago, nicknorman said:

On your last point, the membrane panels are permanently powered if the Nodes are powered. Typically there is a "master switch" input to one of the nodes, and most or all of the services subscribe to it, so that when you turn off the master switch you aren't removing power from the system, you are just forcing everything off except those things you want to remain on such as the bilge pump float switch power. When you press a membrane button to turn something on, the signal is still sent, but nothing happens if/because the virtual "master switch" is off. We use a membrane button for the master switch but I guess in your case it has been configured to use an input from the key switch. There is a buzzer function on the membrane keypad but I've never used it. I do find that when the backlight is on, there is a faint  buzzing/humming sound for some reason.

I suppose another question is, if you disconnect that bad membrane panel, does everything else work exactly as expected?

Tnx for the clear explanation. Good thing, I can rule the main unit out for now. I'm awaiting answer from Sweden - ready for the price tag.

But price is one thing.... I think I need to consider some sort of backup / override function. Maybe a small board with premounted fuses and relays which could be used to secure navigational lights etc in case of failure....  

[nicknorman]

33 minutes ago, Peter Laursen said:

Tnx for the clear explanation. Good thing, I can rule the main unit out for now. I'm awaiting answer from Sweden - ready for the price tag.

But price is one thing.... I think I need to consider some sort of backup / override function. Maybe a small board with premounted fuses and relays which could be used to secure navigational lights etc in case of failure....  

Obviously you have lost some confidence in the system having just bought it and now it has a fault. Admittedly we are on the safety of inland waters, but the system hasn't failed us yet and I don't expect it to. Using MOSFETs to switch nav lights is more reliable than using relays! If you had mechanical switches, fuses and relays to operate the lights from the outset, would you feel the need to add a backup set of switches fuses and relays? Probably not, I suggest! Sorry but IMO your concern is based on fear of the unknown rather than a scientific reliability analysis. The most likely failure modes will be wiring faults, and that could happen regardless of the type of switching.

Bear in mind that the system was designed from the outset for sea-going craft, not narrowboats.

You may think CANBUS and computers are all a bit fancy and complicated, but things like the ABS and Stability Control systems in your car use the same technology.

 

[Peter Laursen]

11 minutes ago, nicknorman said:

Obviously you have lost some confidence in the system having just bought it and now it has a fault. Admittedly we are on the safety of inland waters, but the system hasn't failed us yet and I don't expect it to. Using MOSFETs to switch nav lights is more reliable than using relays! If you had mechanical switches, fuses and relays to operate the lights from the outset, would you feel the need to add a backup set of switches fuses and relays? Probably not, I suggest! Sorry but IMO your concern is based on fear of the unknown rather than a scientific reliability analysis. The most likely failure modes will be wiring faults, and that could happen regardless of the type of switching.

Bear in mind that the system was designed from the outset for sea-going craft, not narrowboats.

You may think CANBUS and computers are all a bit fancy and complicated, but things like the ABS and Stability Control systems in your car use the same technology.

 

 

He he - thanks for the perspective. Confidence in reliability slowly growing! I do get canbus and computers etc. No problem. But I can't fix it with my toolbox and standard parts, if it fails. And certainly my boating partner can't. So I need it to function in it's core - I will take care of corrosion, proper connections and general maintenance 🙂

[Peter Laursen]

Update....
I got the replacement membrane panel yesterday and mounted it today. The old one was indeed the culprit.

As it turns out, I just learned, the the other membrane panel was replaced a couple of months ago by the seller of the boat (done by a prof boat electrician).

Hmmm, so both panels failed within a relatively short period of time.

 

Just as a service info to EmpirBUS-owners - the new panel came with a foam "gasket" and clear instructions, that there MUST be drilled two holes to allow for ventilation and prevention of condensation. See attached. As theese holes are not there in the original installation, it may og may not give a hint as to the course of failure. I don't know. I'll drill the holes and get on with it 🙂

 

Kind regards and thanks for input!

[wakey_wake]

(sorry for digging up a long & old thread, I was looking through and it has some useful info)

  

 

On 15/11/2021 at 10:41, Up-Side-Down said:

[...]  iii) How about simply using a Battery to Battery charger from the start LA battery to the lithium's with a lithium setting?

On 15/11/2021 at 10:44, MtB said:

iii) Yes a B2B would work but introduces a massive 30% or more loss in charging efficiency, in heat chucked out by the electronics, AIUI.

 

Do you have a source for that? (By which I mean, it's not what I expected and I suspect it's nowhere near that bad)

https://sterling-power.com/pages/battery-to-battery-charger-range claims their 4-switch topology is 95%+ which is what I would expect. Holding on to the last few percent is going to get expensive.

 

Asking because I expect to use B2B unless it's LFP in parallel with a small LA (marginal benefits as discussed). I know there are losses, and reducing those means spending more up front and/or being more patient with the B2B transfer.

 

 

On 15/11/2021 at 13:10, nicknorman said:

Modern alternators tend to have zener diodes as the main rectifier diodes. This means that they break down (temporarily and gracefully)at some abnormally high voltage and absorb the transient created with a sudden load dump and before the field current can subside. My Iskra for example has this in the data page:

[... pic... ]

so the diodes should gracefully limit the voltage spike to between 19 and 25v (depending on what the field current was when the load was disconnected, presumably).

Is the response time of zeners fast enough to catch a load dump? Or is the idea for them to absorb a significant fraction of the dump, to spare the TVSs?

They must be chunky ones to carry the forward current, so it makes sense to use them. But I would suspect the transient response could go way up before they start to conduct.

 

Which brings me to this from 2 years ago -

 

On 15/11/2021 at 13:10, nicknorman said:

But no, I haven’t actually tried it! And yes I have just sent away for some small boards from China onto which I’m going to mount some transient absorption diodes. belt and braces!

 

I also have communication between BMS and alternator so not only does the alternator know the individual cell voltages, but if the BMS is going to disconnect on high cell voltage it will give the alternator a few seconds warning and tell it to shut down. Double belt and braces!

 

Thanks for posting the pic at https://www.canalworld.net/forums/index.php?/topic/113208-all-electric/&do=findComment&comment=2713869

I couldn't identify the TVS from "5pes 8ggjd". What is it?

 

Did you "exercise" these protections to see them working? Or was it "install belt, braces and piece of string; then hope for the best" ?

 

I guess load dump protection would have to be tested with a scratch alternator and then a fat relay which switches on and off three times a minute for a while...? Watching with a 'scope or something that's horribly sensitive to over-voltage? An interesting but possibly expensive experiment!

 

 

On 15/11/2021 at 14:48, Up-Side-Down said:

Sorry if I'm being dense (or just plain ignorant) here but please could you explain the relationship between a winter output of 1.5kWh/day and the figure of 60W average you go on to quote?

 

I recommend the tool GNU Units for these sorts of calculations.

 

Home page and introduction: https://www.gnu.org/software/units/

Wikipedia: https://en.wikipedia.org/wiki/GNU_Units

Use it online (without installing to your computer): https://www.fourmilab.ch/webtools/units/ but you're missing out on tab completion, which can be handy.

Explain Like I'm 5: https://eli5.gg/GNU Units

Theory behind this: https://en.wikipedia.org/wiki/Dimensional_analysis

 

It doesn't know the abbreviation "kWh" but it's fine if you spell it "kW hr". It also knows "diesel" as "130500 btu/usgallon".

Here is your question,

 

Convert 1.5 kW hr / 24 hours

To watts

Answer 1.5 kW hr / 24 hours = 62.5 watt

 

What it means is that averaging over those 24 hours, that amount of energy (1.5 electric units) is equal to a steady demand for 62.5 watts for the time period.

 

Watts are joules per second. Kilowatt hours are 1000 joules per second for one hour - so the time component cancels but you're left with a numeric factor which the program keeps track of for you: 1 kW hr  = 3,600,000 joules.

 

If you give it a silly question like "5 miles + 2 hours" then you get an error "Invalid sum or difference of non-conformable units", i.e. you can't do it.

Also if you try to convert "5 miles" into gallons (!) then you get an error. Fair enough.

 

But it can give you "10 km + 50 chain + 1 picolightyear" in any length unit you like.

 

(If you want it to deal with °F then you have to be more careful - let's not get into that)

 

 

  During the bit about efficiency and cost of electricity with various kit,

On 15/11/2021 at 16:57, IanD said:

I don't think you'll get 27%, probably 20% is closer to the mark including all losses so you end up around 40p/kWh -- which is still not too bad, and is half the 80p/kWh of running the engine while moored.

 

LiFePO4 should really be part of the solution though; using LA you not only have the low round-trip efficiency but the need to run for at least a couple of hours regularly to get to 100% SoC. I'd be surprised if these don't put the cost per unit up by al least 50%, probably more than that...

 

there are many disadvantages to main engine + alternator + lead acid, and the need to slowly/eventually reach near 100% SoC to postpone battery replacement costs is a big part of that.

 

Suppose, in the absence of LiFePO4 just to simplify the conversation, one split an LA bank in two. They both get a bulk charge, then stop then engine and use one to power the absorption charge of the other. Possibly you want another short burst of engine to give a second bulk charge to the donor. (Next time, swap them.)

This is a bit weird and I've never seen it, but in theory it prevents sulphation by giving the 100%-ish charge often enough, and not heavily depleting the other half bank to do it. But without spinning the engine at negligible load for pointless hours.

 

I think this makes "LiFePO4 as third bank" look more appealing because it can be the donor for absorption charge, and not be bothered by its lower SOC.

However it doesn't say much to justify keeping the "second" bank of domestic LA, except for the load dump issue in a known-risks way.

 

[Up-Side-Down]

5 minutes ago, wakey_wake said:

(sorry for digging up a long & old thread, I was looking through and it has some useful info)

  

 

 

Do you have a source for that? (By which I mean, it's not what I expected and I suspect it's nowhere near that bad)

https://sterling-power.com/pages/battery-to-battery-charger-range claims their 4-switch topology is 95%+ which is what I would expect. Holding on to the last few percent is going to get expensive.

 

Asking because I expect to use B2B unless it's LFP in parallel with a small LA (marginal benefits as discussed). I know there are losses, and reducing those means spending more up front and/or being more patient with the B2B transfer.

 

 

Is the response time of zeners fast enough to catch a load dump? Or is the idea for them to absorb a significant fraction of the dump, to spare the TVSs?

They must be chunky ones to carry the forward current, so it makes sense to use them. But I would suspect the transient response could go way up before they start to conduct.

 

Which brings me to this from 2 years ago -

 

 

Thanks for posting the pic at https://www.canalworld.net/forums/index.php?/topic/113208-all-electric/&do=findComment&comment=2713869

I couldn't identify the TVS from "5pes 8ggjd". What is it?

 

Did you "exercise" these protections to see them working? Or was it "install belt, braces and piece of string; then hope for the best" ?

 

I guess load dump protection would have to be tested with a scratch alternator and then a fat relay which switches on and off three times a minute for a while...? Watching with a 'scope or something that's horribly sensitive to over-voltage? An interesting but possibly expensive experiment!

 

 

 

I recommend the tool GNU Units for these sorts of calculations.

 

Home page and introduction: https://www.gnu.org/software/units/

Wikipedia: https://en.wikipedia.org/wiki/GNU_Units

Use it online (without installing to your computer): https://www.fourmilab.ch/webtools/units/ but you're missing out on tab completion, which can be handy.

Explain Like I'm 5: https://eli5.gg/GNU Units

Theory behind this: https://en.wikipedia.org/wiki/Dimensional_analysis

 

It doesn't know the abbreviation "kWh" but it's fine if you spell it "kW hr". It also knows "diesel" as "130500 btu/usgallon".

Here is your question,

 

Convert 1.5 kW hr / 24 hours

To watts

Answer 1.5 kW hr / 24 hours = 62.5 watt

 

What it means is that averaging over those 24 hours, that amount of energy (1.5 electric units) is equal to a steady demand for 62.5 watts for the time period.

 

Watts are joules per second. Kilowatt hours are 1000 joules per second for one hour - so the time component cancels but you're left with a numeric factor which the program keeps track of for you: 1 kW hr  = 3,600,000 joules.

 

If you give it a silly question like "5 miles + 2 hours" then you get an error "Invalid sum or difference of non-conformable units", i.e. you can't do it.

Also if you try to convert "5 miles" into gallons (!) then you get an error. Fair enough.

 

But it can give you "10 km + 50 chain + 1 picolightyear" in any length unit you like.

 

(If you want it to deal with °F then you have to be more careful - let's not get into that)

 

 

  During the bit about efficiency and cost of electricity with various kit,

 

there are many disadvantages to main engine + alternator + lead acid, and the need to slowly/eventually reach near 100% SoC to postpone battery replacement costs is a big part of that.

 

Suppose, in the absence of LiFePO4 just to simplify the conversation, one split an LA bank in two. They both get a bulk charge, then stop then engine and use one to power the absorption charge of the other. Possibly you want another short burst of engine to give a second bulk charge to the donor. (Next time, swap them.)

This is a bit weird and I've never seen it, but in theory it prevents sulphation by giving the 100%-ish charge often enough, and not heavily depleting the other half bank to do it. But without spinning the engine at negligible load for pointless hours.

 

I think this makes "LiFePO4 as third bank" look more appealing because it can be the donor for absorption charge, and not be bothered by its lower SOC.

However it doesn't say much to justify keeping the "second" bank of domestic LA, except for the load dump issue in a known-risks way.

 

So I think you've just worked yourself round to a hybrid LA/LiFePO4 set up: lots of YouTube videos out there to expand on the concept, largely dominated by a guy called Clark on a sailing boat called Temptress. He has an axe to grind/product to market (called a Bank Manager) and a lot of what he says is okay (to the point of being almost seductive).

 

However, when the BSS finally gets round to embracing lithium technology – and everything that goes with it – don't be too surprised if hybrid systems of a similar nature get more than frowned upon.

[peterboat]

2 minutes ago, Up-Side-Down said:

So I think you've just worked yourself round to a hybrid LA/LiFePO4 set up: lots of YouTube videos out there to expand on the concept, largely dominated by a guy called Clark on a sailing boat called Temptress. He has an axe to grind/product to market (called a Bank Manager) and a lot of what he says is okay (to the point of being almost seductive).

 

However, when the BSS finally gets round to embracing lithium technology – and everything that goes with it – don't be too surprised if hybrid systems of a similar nature get more than frowned upon.

I have just had my BSS, the inspector was very happy with all my Lifepo4s, fuses and switches kept him content 

[MtB]

19 minutes ago, wakey_wake said:

Do you have a source for that? (By which I mean, it's not what I expected and I suspect it's nowhere near that bad)

https://sterling-power.com/pages/battery-to-battery-charger-range claims their 4-switch topology is 95%+ which is what I would expect. Holding on to the last few percent is going to get expensive.

 

Asking because I expect to use B2B unless it's LFP in parallel with a small LA (marginal benefits as discussed). I know there are losses, and reducing those means spending more up front and/or being more patient with the B2B transfer.

 

No I don't have a source other than having read it stated many times on here, unchallenged so probably true.

 

I think @Tony1 knows more than most on here about B2B chargers and their efficiencies.

 

 

[Up-Side-Down]

10 minutes ago, peterboat said:

I have just had my BSS, the inspector was very happy with all my Lifepo4s, fuses and switches kept him content 

I'd be right in thinking you don't have a LA/LiFePo4 hybrid set up though?

14 minutes ago, MtB said:

 

No I don't have a source other than having read it stated many times on here, unchallenged so probably true.

 

I think @Tony1 knows more than most on here about B2B chargers and their efficiencies.

 

 

Victron quote 89% efficiency for their 30A Orion DC-DC charger.

[peterboat]

13 minutes ago, Up-Side-Down said:

I'd be right in thinking you don't have a LA/LiFePo4 hybrid set up though?

Victron quote 89% efficiency for their 30A Orion DC-DC charger.

I don't, but he knew why people fit hybrid systems, lifepo4s and LAs are a good voltage mix

[MtB]

22 minutes ago, Up-Side-Down said:

Victron quote 89% efficiency for their 30A Orion DC-DC charger.

 

That sounds more like it. Sterling coyly choose not to mention output power divided by input power for theirs, having had a brief scan of their specs.

 

Or maybe I missed it, to be fair.

 

 

 

[Up-Side-Down]

13 minutes ago, MtB said:

 

That sounds more like it. Sterling coyly choose not to mention output power divided by input power for theirs, having had a brief scan of their specs.

 

Or maybe I missed it, to be fair.

 

 

 

I don't think there's ever been a need to be fair to Sterling ......