Cheap LiFePO4 BMS?

[nicknorman]

10 minutes ago, Dr Bob said:

Nick, are you sure of this? My reading of the bumph is that the unit pulls the voltage down from the alternator therefore increasing the alternator output current then increases the voltage to deliver to the batteries.

Now, this is where I really dont understand electrickerty. On LA charging you have bulk where voltage rises, then absorption where you have constand voltage and decaying current etc. On the Li's, when half full, you start the engine and the voltage on the Li bank is 13.5V. If you are charging at less than 50A, this voltage rises until it reaches the point where the AtoB cuts back the current (ie it has reached its absorption phase) dramatically, ie mine will be charging at 35A then all of a sudden it is down to 5A when it gets to 14.0V...ie the US gel setting. What is clear though is that the AtoB stops charging at the fast rate once a target voltage is reached which may or may not be the voltage the alternator regulator is set at. In my case on LAs, the alternator would only deliver 13.9V at constant voltage. If however, I had an alternator regulated at 14.4V, I would expect the AtoB to do exactly the same on the Li's ie charge at increasing voltage until it hits the 14.0V limit (on the US gel setting) and then backs off.

What is certain is that the AtoB is not designed to charge Li's but mine does it very well.

No, definitely not sure! It was something lurking in my mind from earlier discussions on CWDF. I’ve just had a look at the manual and there is a graphs showing output voltage less than alternator voltage so I suspect I’m wrong. Would be worth verifying with Sterling before coughing up a lot of money for one though! You new alternator will regulate at 14.4 to 14.6v

[nicknorman]

7 minutes ago, MoominPapa said:

I'm not anticipating wrecking many A127s. The failure so far is a diode pack on an alternator that's been making most oft he power for full-time liveaboards for five years, and was a cheapy "reconditioned" unit from Ebay when "new".  The experience for a 70A alternator charging a 600Ah LA bank is not significantly different to charging a LiFePO4 bank: in both cases is spends considerable time (hours) turned up to 11.

 

My alternator is at the top of an engine in a well ventilated engine room, which helps. The other thing which helps is that I dismantle it once a year and clean out the substantial dust and fluff accumulations in the air passages. That makes a big difference and I'd recommend it to all as a maintenance practice.

 

MP.

  

Good point but I suggest that it is because your alternator is effectively in a living space that gives it both better cooling and a ready supply of fluff and dead skin cells! I didn’t really find anything more than a bit of black belt dust when I took ours to bits after 9 years, it lives under boards and the space is externally ventilated.

Edited May 13, 2020 by nicknorman

[Dr Bob]

25 minutes ago, ivan&alice said:

It's a Beta 38 which has a serial number that has worn off, I have "declaration of conformity" documents that specify my engine serial number as WY5039 so I presume it is exactly as old as the boat - 21 years old in November. The engine runs like a dream though - with 5-10 seconds of glow plugs prior, it has started twice a day every day for two years, in the coldest winter and hottest summer, with neither incident nor complaint.

I would love to know what my alternator is. If you have any suggestions of how I could find out, I'm all ears!

The declaration of conformity specifies a 60A, 12V alternator, so I'm guessing that this is the "engine paint blue" secondary alternator (10cm in diameter), rather than the silver larger alternator (15cm in diameter). 60A is not to be sniffed at actually - so if I did go for a combined approach that would add up to 130A total, if my other alternator is 70A - but actually just based on the physical size difference perhaps it is a 90A actually.

Could I use my pair of alternators with a single larger B2B or would this be verboten?

 

Otherwise, I'd have to design my basic alternator controller to monitor the voltage of the batteries as well as the temperature of both alternators and shut them off independently.

I think what I could do in the meantime is see what amperage I can get out of my smaller alternator?

 

Here are some pics of my engine and alternators, perhaps someone can enlighten me - I can also take things apart and send more pics if that would be helfpul.

The little one looks like mine ie 50A. The big one is different and looks smaller than my 90A domestic (on the Beta 43 of same age) so maybe is 70A. Someone on here might recognise it.

The AtoB is specifically designed to feed from 2 alternators and feed back to 2 sets of batteries where the engine start is prioritised at the start and then every 20 mins. I cant see how the BtoB could be fed by both alternators....but can you charge the engine start battery from both alternators and then use the BtoB to transfer the power? Not a clue. Above my pay grade. I am sure Nick or MP will know.

[MoominPapa]

53 minutes ago, nicknorman said:

Good point but I suggest that it is because your alternator is effectively in a living space that gives it both better cooling and a ready supply of fluff and dead skin cells! I didn’t really find anything more than a bit of black belt dust when I took ours to bits after 9 years, it lives under boards and the space is externally ventilated.

Also good point.

 

MP.

 

[nicknorman]

1 hour ago, Dr Bob said:

The little one looks like mine ie 50A. The big one is different and looks smaller than my 90A domestic (on the Beta 43 of same age) so maybe is 70A. Someone on here might recognise it.

The AtoB is specifically designed to feed from 2 alternators and feed back to 2 sets of batteries where the engine start is prioritised at the start and then every 20 mins. I cant see how the BtoB could be fed by both alternators....but can you charge the engine start battery from both alternators and then use the BtoB to transfer the power? Not a clue. Above my pay grade. I am sure Nick or MP will know.

For a B2B and 2 alternators I think you would just need some relay / contactor / isolator switch to connect the positives of the leisure and engine battery together when the engine was running. That way, both alternators would feed the B2B.

 

Downside that I can think of off the top of my head would be that the alternators would have to have regulated voltages reasonably close, otherwise you could end up with one at full output and getting too hot, whilst the other languished. But probably if both were fairly modern and working at 14.4 v nominally, it would be fine. For the sake of a battery isolator and some cable, worth trying perhaps?

[WotEver]

3 hours ago, Dre said:

If by that you mean a "load dump" (great name ?)...

No, I believe he was referring to what Nick called a 'freewheel diode', which I'd call a 'flyback diode', either of which is a very basic 'snubber'.

 

It's simply a reverse-biased diode across the relay contacts.

 

 

 

Edited May 13, 2020 by WotEver
the dreaded duplication bug

[WotEver]

2 hours ago, ivan&alice said:

I think what I could do in the meantime is see what amperage I can get out of my smaller alternator?

Current!

 

Don't be an American, please

[nicknorman]

8 minutes ago, WotEver said:

No, I believe he was referring to what Nick called a 'freewheel diode', which I'd call a 'flyback diode', either of which is a very basic 'snubber'.

 

It's simply a reverse-biased diode across the relay contacts.

 

No in this case it is not across the relay contacts, it’s across the rotor coil (normally reverse biased, as you say).

Edited May 13, 2020 by nicknorman

[WotEver]

Just now, nicknorman said:

No in this case it is not across the relay contacts, it’s across the rotor coil (normally reverse biassed, as you say).

Oh yeah, of course.

[jetzi]

 

6 hours ago, Dr Bob said:

as I say above and bypass it letting more Amps in and controlling only on alternator temp for the autodisconnect relay.

Arg but this would then necessitate both an alternator control system and a B to B.

 

6 hours ago, nicknorman said:

I’m pretty sure that important point you are missing is that an A2B’s purpose is to boost the alternator voltage using a step-up device. It can’t make the output voltage less than the alternator’s own regulator, it can only make it more.

 

Whereas the B2B is step up or step down device, which is why it can go into “float”.

An A2B is then useless in this application then - I only need to step down voltage when 80% SoC is reached, not step up.

 

6 hours ago, nicknorman said:

Oh and on the freewheel diode thing, below is an extract from the internal circuitry of the chip im using, which shows it. The 3 connections are the B+A which would be the D+ terminal in your case, the EXC (excite) which is the connection to the rotor brush, and ground. If you replace the MOSFET with your relay, and add the diode as shown, rated at say 10A (well, 5A minimum) then all will be sweet.

I'm not sure I understand the circuit diagram, but I think I somewhat understand the freewheel diode concept. Essentially we wouldn't want to turn off the field without providing some route for the inducted current to continue flowing, otherwise the relay could arc and damage the contacts. Would the "earth" be the battery negative (engine block), the hull, or what?  @Sir Nibble established in another thread that my alternator is self-exciting so I won't have that connection, so I'm still not sure how that circuit diagram would translate to my particular situation.

 

6 hours ago, Dr Bob said:

The problem with mine is that if both alternators are fed into the AtoB it only takes power from the small one as the voltages each puts out are different. The 50A produces 14.3V and the 90A only 13.9V. I bought the AtoB when I had LAs and the 13.9V was not high enough to charge them (see Gibbo's comments on crap old automotive alternators - I have one!)

With my engine running I measure 14.65 volts across both alternators, but I could be measuring the same thing here - I'm not entirely sure how the small alternator is wired at the moment, it's output wire disappears into an engine wiring conduit. Perhaps the first thing I need to do is work out exactly how it's currently wired. Next thing I think would be to disconnect my MPPT, and wire the alternators individually in turn to a load and see what current and voltage they put out.

 

5 hours ago, MoominPapa said:

The other thing which helps is that I dismantle it once a year and clean out the substantial dust and fluff accumulations in the air passages. That makes a big difference and I'd recommend it to all as a maintenance practice.

Mine is full of diesel soot (yes diesel soot, not rubber dust) that I don't really know how to get out. It needs to be thoroughly washed I think but introducing water into an alternator has got to be one of the worst ideas I could think of!

 

3 hours ago, nicknorman said:

Downside that I can think of off the top of my head would be that the alternators would have to have regulated voltages reasonably close, otherwise you could end up with one at full output and getting too hot, whilst the other languished. But probably if both were fairly modern and working at 14.4 v nominally, it would be fine. For the sake of a battery isolator and some cable, worth trying perhaps?

What would trying look like if you didn't yet have a B2B - would it be possible to get a good idea if it would work before buying? If I were to flatten and then disconnect my leisure battery, and hook it up to just one alternator, just the other, then both together, could I get a good idea of how much power the alternator can producing, what voltage each was running at and whether the voltages were close enough to be used as a combined input for a B2B?

 

2 hours ago, nicknorman said:

No in this case it is not across the relay contacts, it’s across the rotor coil (normally reverse biased, as you say).

 

Just to confirm, am I right in thinking basically the diode just completes the alternator coil (in the "wrong" direction) so that when the coil is turned off, the coil forms a circuit with itself and the diode?

 

(alternator output and relay control not shown)

 

 

2 hours ago, WotEver said:

Current!

 

Don't be an American, please

How terribly embarrassing! It just slipped out... I do apologise!

Edited May 13, 2020 by ivan&alice

[Sir Nibble]

The "alternator coil" in the diagram is the rotor, the field coil. The freewheel diode already exists in the regulator circuitry.

Purely incidentally, the diode in the diagram, as drawn, is a short circuit.

Edited May 13, 2020 by Sir Nibble

[jetzi]

5 minutes ago, Sir Nibble said:

The "alternator coil" in the diagram is the rotor, the field coil. The freewheel diode already exists in the regulator circuitry.

Perhaps the concern is that if we dissect the alternator and insert a relay to turn the field coil off to prevent LiFePO overcharge / alternator overheat, we're bypassing the regulator?

[Sir Nibble]

9 minutes ago, ivan&alice said:

Perhaps the concern is that if we dissect the alternator and insert a relay to turn the field coil off to prevent LiFePO overcharge / alternator overheat, we're bypassing the regulator?

There has to be regulation so therefore the regulator, or A regulator will be connected, they all incorporate the diode or the inductive back emf would instantly destroy the switching transistor. So long as it is either the supply or return to the regulator that is interrupted rather than the brush connection then no problem.

[Dr Bob]

34 minutes ago, ivan&alice said:

 

 

 

An A2B is then useless in this application then - I only need to step down voltage when 80% SoC is reached, not step up.

 

No, see my response to Nick above. Voltatge increases on the AtoB until the target voltage is reached. These are Li's....totally different from LAs

My AtoB works very very well but may be out of your budget.

[WotEver]

40 minutes ago, Sir Nibble said:

The "alternator coil" in the diagram is the rotor, the field coil. The freewheel diode already exists in the regulator circuitry.

Purely incidentally, the diode in the diagram, as drawn, is a short circuit.

@Sir Nibble how easy/difficult is it to access the field connection for this purpose? Or does it entirely depend on the alternator?

[Sir Nibble]

21 minutes ago, WotEver said:

@Sir Nibble how easy/difficult is it to access the field connection for this purpose? Or does it entirely depend on the alternator?

It does depend. On many American alternators the connections are external with the exception of the brush connections. On an A127 it's that yellow lead from the field diodes. Bosch alternators would need careful surgery to the alternator. With a nine diode machine the easiest way is to slap a -ve onto D+.

This is an American motorola unit and opening the red lead would do the job. 

 

 

Japanese stuff is just way too varied and may need internal surgery or may be done externally depending. Give me a picture of the machine in question and I will advise if I can.

 

 

 

 

 

 

 

Edited May 13, 2020 by Sir Nibble

[nicknorman]

57 minutes ago, Sir Nibble said:

There has to be regulation so therefore the regulator, or A regulator will be connected, they all incorporate the diode or the inductive back emf would instantly destroy the switching transistor. So long as it is either the supply or return to the regulator that is interrupted rather than the brush connection then no problem.

Yes there is a freewheel diode in the regulator of course. But if you want to shut down the alternator you have to break into the circuit to introduce a relay. I think it will depend on where you break into it, but an additional diode would be required if you break into it between the reg and the stator.

[Sir Nibble]

21 minutes ago, nicknorman said:

Yes there is a freewheel diode in the regulator of course. But if you want to shut down the alternator you have to break into the circuit to introduce a relay. I think it will depend on where you break into it, but an additional diode would be required if you break into it between the reg and the stator.

Well I can't see it. The diode is across the brushes and if the regulator is still connected to them then the diode is in place. If the supply to the reg from either field diodes or battery is opened then that's it.

[nicknorman]

50 minutes ago, Sir Nibble said:

Well I can't see it. The diode is across the brushes and if the regulator is still connected to them then the diode is in place. If the supply to the reg from either field diodes or battery is opened then that's it.

Yes I agree. But, depending on the type of alternator, it might be easier to introduce the relay between the regulator and the rotor. I’m just thinking of my Iskra which has the brush and reg unit as a module, the field is connected to the reg/brush module via a screw down lug and it wouldn’t be that easy to retain a connection to both the field and the reg, but spilt them to insert the relay contacts. It would be much easier to snip the wire linking the reg to brush and insert it there, along with a diode of course!

 

Older types of alternator I’m sure it would be best to break the connection between the field diodes and the reg, and then the reg continues to provide the rotor with its built in freewheel diode.
 

I think we are on the same page, I was just sounding an alert that to break the connection between the reg and the rotor without a freewheel diode, would be a really bad idea.

 

On mine, it is the green screw in the pic that transfers the current from the field diodes to the reg module. You’d have to bend it up enough to get the screw back in without going through the lug, and connect a separate wire to the lug.

 

 

 

 

Edited May 13, 2020 by nicknorman

[nicknorman]

Anyway, good discussion and it’s made me realise I need another switch on my regulator to select “slow charge” for when we will be cruising all day, to maximise battery and alternator life, and “fast charge” for when we are static and want to top up as fast as possible.

Edited May 14, 2020 by nicknorman

[nicknorman]

On 13/05/2020 at 21:54, nicknorman said:

I’m just thinking of my Iskra which has the brush and reg unit as a module, the field is connected to the reg/brush module via a screw down lug and it wouldn’t be that easy to retain a connection to both the field and the reg, but spilt them to insert the relay contacts. It would be much easier to snip the wire linking the reg to brush and insert it there, along with a diode of course!

Just realised the above doesn’t make sense, too late to edit the post. It should have said:

 

“I’m just thinking of my Iskra which has the brush and reg unit as a module, the field diode supply is connected to the reg/brush module via a screw down lug and it wouldn’t be that easy to retain a connection to both the field diode supply and the reg, but spilt them to insert the relay contacts. It would be much easier to snip the wire linking the reg to brush and insert it there, along with a diode of course!”

[jetzi]

Plan for this weekend is to test both my alternators and get some real data to work with. I'll disconnect them from everything except my test setup: a 50% SoC LA battery plus a 600W 12V (50A) immersion heater element.

I'll measure voltage and current, for each alternator individually and both together.

 

If the two alternators are very close in voltage, I can get a B2B that is no more than 80% of the combined measured current once they are hot. The practical current is likely to be significantly less than what the alternators are theoretically rated for, but it's actually what matters when sizing the B2B. If, for example, I'm getting only 40-60A in total from my alternators, I'll get the 30A Sterling B2B refurb and call it a day - it's cheap and simple. If I'm getting 70-90A total, I'll consider the 60A Sterling B2B - but the price point then becomes a big issue for me. >300 GBP seems a lot for a solution that's less flexible and less efficient than a less costly DIY alternator controller.

 

If the alternators are NOT well within 0.2V or so, then it's back to the alternator surgery + custom DIY controller system. Mine would be much more basic than Nick's - as nice as it would be, I don't think I'll bother with regulating a float voltage. Instead I'd just make the arduino shut down the field and hence charge (incorporating a freewheel diode) on either high voltage or high temperature.

 

I'm still not totally convinced I shouldn't just rely on the BMS+relay shutdown, incorporating a LA dump load on the alternator side. I realise it is one safety layer fewer, but as several people have pointed out they've never needed more than one layer of safety so far, I'll still have the other 3, and AFAICT routinely shutting off charge really is the intended use of the BMS. The BMS I bought has an overvoltage shutoff driving a relay on the charge, an undervoltage shutoff driving a relay on the load, AND an emergency shutoff that can be used to drive a third relay on the negative for both under and overvoltage, so I still have that emergency relay shutoff. I hear the concern that my Tycos are going to have a shorter life if they are used routinely, but

1. firstly I'm working with a smaller alternator than most here,
2. secondly I'll try to manually avoid overcharging (by monitoring voltage, running engine only when flat, manually tripping the alternator charge relay if I'm going on a cruise and my LiFePOs are full, and oversizing the battery bank), and
3. thirdly I bought a couple of spare relays - I'm willing to melt one to see how long it takes to do so.

 

Alternator charging of LiFePOs - definitely the trickiest part of the lithium project, but I feel like I have a way forward...

Edited May 15, 2020 by ivan&alice

[nicknorman]

Im a bit sceptical about your expected current outputs. In my experience alternators can put out their rated output provided they are spinning fast enough, until they get very hot at which point the output decreases a bit. I’ve seen 180A from our 175A alternator when it’s cold. When it’s hot, perhaps only 165A.

 

I think the main reason why people feel their alternators don’t produce their rated output is because they aren’t loaded adequately. The regulators have a very “soft” curve such that full output is only obtained when the terminal voltage is quite a bit below the regulated value, maybe a volt or so. If the voltage rises towards the regulated value, the current falls off a lot, long before the regulated voltage is reached. If your charging circuit drops 0.25v in the +ve connections and cable, 0.25v in the -ve connections and cable, and 0.25vin the battery isolator, your alternator will be 0.75v above battery voltage and the current will thus be quite low.

 

I suspect you may find that with Li batteries at a low SoC and presuming you have decent low resistance cabling, that your alternators can produce their rated outputs. Which is probably not what you want from a overheating perspective!

 

If your 90A alternator really can only produce 50A, I would suspect a blown diode.

 

[Dr Bob]

8 hours ago, ivan&alice said:

secondly I'll try to manually avoid overcharging (by monitoring voltage, running engine only when flat, manually tripping the alternator charge relay if I'm going on a cruise and my LiFePOs are full, and oversizing the battery bank), 

 I think you will find this bit very easy. As the Li bank is likely most of the time to be between 20 and 80%, one glance at the BMV to see how many amphrs out(or when you get more experience just the voltage and current out ) each morning and a look out the window at the sun and you can decide if you isolate the Li's from the alternator for the day or how may hours you will run the engine for the day. 

When designing a bank, you can spend hours overthinking all of this and everything just falls into place when you actually do it. I've heard enough that you know what you are doing and so you are not going to wreck them. You will get into a routine and everything will be very very simple. Once in that routine they are so much simpler than LAs.

 

Edited May 16, 2020 by Dr Bob

[redwing]

8 hours ago, ivan&alice said:

 If I'm getting 70-90A total, I'll consider the 60A Sterling B2B - but the price point then becomes a big issue for me. >300 GBP seems a lot for a solution that's less flexible and less efficient than a less costly DIY alternator controller.

 

 

Ivan, Sterling do have some 60a refurbs @ £200, And there are some new @ 240 on ebay

 

They also have the half power function that Dre mentioned the other day. Would be good to hear how he is getting on