48v alternators

[IanD]

2 minutes ago, Antrepat said:

I've had some response from Beta.  I told them I wanted a quote for a replacement for my existing engine (14hp) which would deliver more power and would be suitable to convert in the future from driving the propeller to driving only a large alternator for battery charging.  They've come back recommending a Greenline 20 engine (about £6k inc. VAT), with the caveat that it would need to spin at at least 2,200rpm to reach a sufficient point on the power curve to drive the alternator I indicated (that Mahle one).  So far they haven't said they foresee any drive belt etc. issues with hooking up a much (much) larger alternator, but I'm working on focussing their minds on the issue, asking specifically about whether it's a problem and whether mounting it off the flywheel would be better.

 

I've followed up with an observation that their 2/6 genset uses a really compact 2-cylinder engine and is rated for 6-7kw at 3000rpm.  They don't do a 48v generator, but I've asked whether they foresee any issue getting the same engine (Kubota BZ482) and attaching the Sincro E3 2-48/150 (150A, 2-pole, 48v, 3000rpm) rather than Beta's AC generator.

By the time you've bought a new engine like the Greenline 20 a big alternator and the controller for it and mounted everything and soundproofed it, you'll probably have spent as much as buying a generator -- maybe more... 😉

[cuthound]

2 minutes ago, IanD said:

By the time you've bought a new engine like the Greenline 20 a big alternator and the controller for it and mounted everything and soundproofed it, you'll probably have spent as much as buying a generator -- maybe more... 😉

 

I agree.

 

A cheaper alternative might be to buy a secondhand generator set and noise insulate it 

 

BT often sell their ex-standby generators in the 10-25kVA range with maybe 150 hours on them. They are available from brokers (check the internet) for £4-5000.

 

Transform the output down to 48 volts and Robert is your father's brother.

[Antrepat]

2 minutes ago, IanD said:

By the time you've bought a new engine like the Greenline 20 a big alternator and the controller for it and mounted everything and soundproofed it, you'll probably have spent as much as buying a generator -- maybe more... 😉

Greenline 20 - £6,000

Mahle MG93 - £1,290

Wakespeed WS500 - £475

Connection gear - £500?

= £8,500 or so

 

Soundproofing?? Haven't got any at the moment.  Could line the engine bay with loft insulation, maybe.  (Not supposed to be a serious suggestion.)

 

How much would an 8kw marine generator be new, then?  (As for second hand, I haven't found much but maybe I'm not finding the right sites.)

 

Here's an alternative:

Kubota BZ482 - £4,000?

Sincro E3 2-48/150 - £850

(Controller) - £500

(Connection gear not required: mates axially, comes with fittings.)

= £5,500

 

...but then I'd have to go electric immediately...

[Antrepat]

2 minutes ago, cuthound said:

A cheaper alternative might be to buy a secondhand generator set and noise insulate it 

 

BT often sell their ex-standby generators in the 10-25kVA range with maybe 150 hours on them. They are available from brokers (check the internet) for £4-5000.

 

Transform the output down to 48 volts and Robert is your father's brother.

 

As it happens, Robert is my dad.  David's me father's brother.

 

I'd be perfectly happy to get a secondhand generator, if I could find anywhere to get one.  There are loads of dealers in the kinds of generators you might use on a building site, sure, but marine ones?  I'll try to find these brokers, but a few links would be helpful...

 

I don't have a noise-insulated engine at the moment but I would like to add something if I can find any advice on what.

 

Transform, rectify, 10kVA, that's pretty serious power.  With what?  And how about control?  LFP charging needs prompt charge termination.  Something like the Wakespeed WS500 is designed to control alternators directly, but with a generator, I suppose the BMS would need to sense the termination voltage/current conditions and send an auto-stop signal to the generator.  That sort of integration was what I could only dream of...

[Antrepat]

20 minutes ago, cuthound said:

A cheaper alternative might be to buy a secondhand generator set and noise insulate it

 

I just had a quick look around and that'd work if I wanted to power a hospital or something.  Most of what's readily visible seems to be of that scale.  The idea of then figuring out what I'd need to transform and rectify to 48v DC is daunting, and in any case, none of these generators is "marinised", doesn't that matter?

[IanD]

6 minutes ago, Antrepat said:

 

I just had a quick look around and that'd work if I wanted to power a hospital or something.  Most of what's readily visible seems to be of that scale.  The idea of then figuring out what I'd need to transform and rectify to 48v DC is daunting, and in any case, none of these generators is "marinised", doesn't that matter?

 

If you have an AC generator the normal way to connect this to the batteries is via something like a Victron Multiplus/Quattro, which acts as a battery charger and an inverter. This also has the benefit that if the inverter dies you've still got a mains source (direct from generator). And as I'm sure you've found out it's *far* easier to get 230Vac generators (massive market, many models from many suppliers including cheaper ones) then 48V DC ones (tiny market, few models from few suppliers, none cheap). But then a Multiplus II 48/10000 (which you need to charge batteries at 7kW) is advertised at £3400, assuming this includes the latest Victron price rises (which it may not do), so again not cheap...

 

Non-marinised generators (e.g. ex-hospital or standby) tend to use radiators and fans (and a *lot* of air) to keep them cool, which is not convenient on a boat where you really want water cooling --- and for the canals, keel cooling not freshwater. It may be possible to convert them with some difficulty, or may not be. Most marine generators are freshwater-cooled (intended for sea use with a pumped water inlet/output), only a few (e.g. Beta Marine) are available with keel cooling as standard -- though some (e.g. Fischer-Panda, Kohler) have kits to do this buried away in the manufacturers parts list.

[Antrepat]

1 hour ago, IanD said:

 

If you have an AC generator the normal way to connect this to the batteries is via something like a Victron Multiplus/Quattro, which acts as a battery charger and an inverter. This also has the benefit that if the inverter dies you've still got a mains source (direct from generator). And as I'm sure you've found out it's *far* easier to get 230Vac generators (massive market, many models from many suppliers including cheaper ones) then 48V DC ones (tiny market, few models from few suppliers, none cheap). But then a Multiplus II 48/10000 (which you need to charge batteries at 7kW) is advertised at £3400, assuming this includes the latest Victron price rises (which it may not do), so again not cheap...

 

Non-marinised generators (e.g. ex-hospital or standby) tend to use radiators and fans (and a *lot* of air) to keep them cool, which is not convenient on a boat where you really want water cooling --- and for the canals, keel cooling not freshwater. It may be possible to convert them with some difficulty, or may not be. Most marine generators are freshwater-cooled (intended for sea use with a pumped water inlet/output), only a few (e.g. Beta Marine) are available with keel cooling as standard -- though some (e.g. Fischer-Panda, Kohler) have kits to do this buried away in the manufacturers parts list.

 

Skin tank cooling, only practical way.

 

I think DC generators are priced so high because people are prepared to pay that.  It's not justified, as far as I can tell, by the cost of the components or the mechanical or electrical complexity of the rig.  Basically, an engine driving an x kVA 230v AC alternator needs to do the same amount of work as an x kVA 48v DC alternator, doesn't it?  It burns fuel, converts (some of) the energy into rotary motion, which does mechanical work to force the rotor through the magnetic field, causing potential to arise and charge to flow.  The difference is in the windings of the alternator and whether the output is rectified or not, not in the amount of work needing to be done by the engine.  If I'm missing something about the Physics here, someone please say.

 

Evidently you can get a 48v DC alternator outputting 150A for maybe £1,300, including an axially-mounted one.  It is patently not the case that a £6,000 (+VAT) AC generator suddenly needs to be a £13,000 (+VAT) DC generator.  They're charging what people seem to be willing to pay.  I know there will be economies of scale to making 1,000 AC generators a year rather than 10 DC ones, but we don't need to pay that premium, not if it's a matter of bolting a different type of alternator to the same engine in the same way to do the same amount of work.

 

The Beta 2/6 genset uses a Kubota BZ482 2-cylinder engine to deliver 6.5kVA 50Hz 230v AC at 3000rpm.  The Sincro 48v DC 100A 2-pole generator is currently available for about £850 and delivers its goods going at 3000rpm.  Set to output 58v, that would be 5.8kVA at full tilt, or 8.7kVA for the 150A version (£1,280).  (Sadly there isn't a 125A 2-pole).  Get a Kubota BZ482 (which has got to be cheaper than a Beta Greenline 20, hasn't it?), mount a Sincro on it, and get a 6kVA 48v DC generator (set to 58v) for maybe under £6k (including bloody VAT).  Why is this too good to be true?

 

Then the issue reverts to being one of control: how do you get the generator to stop, gracefully and promptly, once a predetermined charge termination current and voltage have been reached?

[nicknorman]

1 minute ago, Antrepat said:

 

S If I'm missing something about the Physics here, someone please say.

 

 

 

Low voltage power generation is more difficult than high voltage, at least in terms of amount of copper needed, cooling, efficiency etc. I cite my 12v 175A Iskra alternator that gets extremely hot when it is producing 175A (about 2.2kw) whereas the Travelpower, which is 3.5kw at about 300v ac doesn't seem to struggle, even though it is if anything physically a bit smaller.

[Tony Brooks]

10 minutes ago, Antrepat said:

 

Then the issue reverts to being one of control: how do you get the generator to stop, gracefully and promptly, once a predetermined charge termination current and voltage have been reached?

 

In the case of DC generators or AC alternators the answer is the same. Open circuit the field coils or stator. A simple relay would do that in response to whatever you are using to monitor the charge.

 

More difficult on a permanent magnet machine but they must have some form of voltage control that could be utilised to drop the output voltage below the battery voltage.

Edited October 17, 2022 by Tony Brooks

[Antrepat]

5 minutes ago, nicknorman said:

 

Low voltage power generation is more difficult than high voltage, at least in terms of amount of copper needed, cooling, efficiency etc. I cite my 12v 175A Iskra alternator that gets extremely hot when it is producing 175A (about 2.2kw) whereas the Travelpower, which is 3.5kw at about 300v ac doesn't seem to struggle, even though it is if anything physically a bit smaller.

You're right: the objective is to supply energy at the same rate, and P = VI so less V, more I for the same P.  That's the same inside the alternator as anywhere, so its windings need to carry much higher currents, which means thicker wires.  Why would it be necessarily less efficient, though (get hotter, waste more energy as heat)?  Do the windings have to be more marginal to save weight (or expensive copper), and thus get hotter?

 

The Sincro is quoted as between 75% and 85% efficient, depending on the exact model, at 3/4 load.  This implies it is better to get a slightly higher powered one as I'm thinking its efficiency peak will be at that kind of load: better to operate a 150A unit at 3/4 power than a 100A at full power.  This efficiency is a conversion rate, which in turn shows the engine power needed and implies a heat output of (let's be conservative, though I expect "exact model" is the difference between the 24v and 48v ones) 25% of the input power.  If the output power is 3/4 of full power, that would be 112.5A at 58v: 6.525kW.  That's the 75%, so the heat would be 2.195kW and the input 8.720kW - almost 12hp.  That heat needs to be dissipated in addition to the heat from the engine.  Anyone want to guess what the heat output of an engine delivering 12hp of effort will be?  (Assuming the engine is properly specified and is not having to work excessively hard.)

 

[BEngo]

The rule of thumb is that 1/3 of the input diesel energy comes out of the flywheel, belts etc.as usable work, 1/3   of the input energy goes up the exhaust and the remainder goes into the cooling air/water.  That obviously ignores losses to mechanical noise,  and radiated or convected heat from the block of a water cooled engine.

 

So 12hp at the flywheel equals about 9 or 10 kW to dissipate from the cooling medium.

 

N

[Antrepat]

23 minutes ago, Tony Brooks said:

 

In the case of DC generators or AC alternators the answer is the same. Open circuit the field coils or stator. A simple relay would do that in response to whatever you are using to monitor the charge.

 

More difficult on a permanent magnet machine but they must have some form of voltage control that could be utilised to drop the output voltage below the battery voltage.

 

The datasheet for these Sincro units I'm on about says: "Adjustable voltage set-point by trimmer, or external potentiometer, or outer controller" (dodgy translation) and "Operating output voltage (settable) From 48 VDC to 58 VDC".  Here's the manual - English from page 16, regulation information on page 24, such as it is.  I don't think this device allows direct control of the field coils, which of course is what external regulators like the Wakespeed are designed to do, but anyway.  Reducing the output to below about 54.4v would end charging, but something needs to send that signal, calibrated to whatever the right voltage would be to achieve that output, followed preferably by an engine shut-off instruction or at least something to set off an alert that charging is complete and the engine can be shut down.  To me the obvious device would be the BMS, since that constantly monitors the voltage, current, and temperature of the batteries; but which BMSs are capable of outputting a suitable signal?

[Antrepat]

2 minutes ago, BEngo said:

The rule of thumb is that 1/3 of the input diesel energy comes out of the flywheel, belts etc.as usable work, 1/3   of the input energy goes up the exhaust and the remainder goes into the cooling air/water.  That obviously ignores losses to mechanical noise,  and radiated or convected heat from the block of a water cooled engine.

 

So 12hp at the flywheel equals about 9 or 10 kW to dissipate from the cooling medium.

 

N

 

Well, 2.something kw is only adding about 20% then, but still, I'd want to think carefully about keeping my engine bay cool.  It was pointed out to me on here ages ago that an engine aspirates huge amounts of air (I vaguely recall 70 litres a minute or something) so there is always a big draw of fresh air in, and hence forced-draught ventilation is not usually required.

[BEngo]

For each litre of a 4 stroke engine capacity, when running at 1500 rpm,  something approaching 750 litres of air is drawn in every minute.  The exact number depends on volumetric efficiency.

 

N

[Antrepat]

2 hours ago, BEngo said:

For each litre of a 4 stroke engine capacity, when running at 1500 rpm,  something approaching 750 litres of air is drawn in every minute.  The exact number depends on volumetric efficiency.

 

N

I was a factor of ten too low, then.  That kind of throughput will shift a lot of heat.

[IanD]

4 hours ago, nicknorman said:

 

Low voltage power generation is more difficult than high voltage, at least in terms of amount of copper needed, cooling, efficiency etc. I cite my 12v 175A Iskra alternator that gets extremely hot when it is producing 175A (about 2.2kw) whereas the Travelpower, which is 3.5kw at about 300v ac doesn't seem to struggle, even though it is if anything physically a bit smaller.

Low voltage (e.g. 12V) alternators lose a lot of efficiency in the rectifier diodes -- typically these have >2V drop, so maybe 15% power loss. This drops to about 8% for 24V, 4% for 48V, and a negligible amount for a Travelpower. In theory the winding loss is the same (with the same amount of copper), shorter thicker wires for low voltage have the same power loss as longer thinner wires for high voltage.

 

But at 12V they need very thick winding wire and there's a temptation to skimp on this because it's expensive, and in a (cheap) car alternator nobody cares if the losses are high. Travelpower are a lot more expensive and higher voltage so the losses are lower, IIRC they're about 75% efficient compared to about 50% for a 12V alternator -- 48V alternators are a little bit worse than HV, maybe 70%.

[Antrepat]

2 minutes ago, IanD said:

Low voltage (e.g. 12V) alternators lose a lot of efficiency in the rectifier diodes -- typically these have >2V drop, so maybe 15% power loss. This drops to about 8% for 24V, 4% for 48V, and a negligible amount for a Travelpower. In theory the winding loss is the same (with the same amount of copper), shorter thicker wires for low voltage have the same power loss as longer thinner wires for high voltage.

 

But at 12V they need very thick winding wire and there's a temptation to skimp on this because it's expensive, and in a (cheap) car alternator nobody cares if the losses are high. Travelpower are a lot more expensive and higher voltage so the losses are lower, IIRC they're about 75% efficient compared to about 50% for a 12V alternator -- 48V alternators are a little bit worse than HV, maybe 70%.

 

OK, so efficiency of 75%-85% is plausible then.

[IanD]

Just now, Antrepat said:

 

OK, so efficiency of 75%-85% is plausible then.

85% is unlikely, suggest you go and dig into the Travelpower data sheets. AC generator losses are a bit lower then DC because there are no diodes and no AC-DC conversion or inversion, the generator produces 230Vac directly from the windings. But then you have the losses in the AC-DC conversion in the charger so total efficiency ends up similar...

[Antrepat]

2 minutes ago, IanD said:

85% is unlikely, suggest you go and dig into the Travelpower data sheets. AC generator losses are a bit lower then DC because there are no diodes and no AC-DC conversion or inversion, the generator produces 230Vac directly from the windings. But then you have the losses in the AC-DC conversion in the charger so total efficiency ends up similar...

That's why I used 75% for the calculation

[IanD]

10 hours ago, Antrepat said:

That's why I used 75% for the calculation

I used 70% for the 5kW travelpower, which IIRC came from the data sheet figures (it's not stated directly, you have to work it out) -- this includes the losses in the internal rectifier and DC-AC inverter.

 

The Balmar 48V 100A alternator is similar, but they do give figures (see attached), it varies from 72% at 3000rpm (alternator speed, pulley ratio means engine rpm is 2.77x lower) to 67% at 8000rpm.

 

The Iskra 24V alternator came out as about 60% and the 12V one about 50% due to higher diode losses.

 

So 75% looks optimistic, and 85% even more so... 😉

 

If you want to compare all this to a conventional off-the-shelf 230Vac generator, these usually come out very close to 25% overall efficiency (see Victron generator report). At 1500rpm the diesel will be very close to 30% efficient (plot for Beta 43 attached since I have this), which puts the internal 230Vac alternator efficiency at about 83% (no diode losses).

 

BTW I agree with you to some extent about the cost of 48V DC vs. 230Vac generators, some of the higher price is due to smaller volumes, less competition, and "what the market will pay" -- like 48V bow thrusters, as I've found. But you're wrong saying that the cost is equivalent, because with an AC generator there's just the 2-pole/4-pole (3000/1500rpm) alternator producing AC directly, but the "DC generator" either uses a 3-phase alternator plus rectifier diodes to generate high-voltage variable DC (like the Travelpower) and a DC-DC converter (cost, loss) to generate 48V from this, or maybe a 6-phase alternator with field coil regulation and rectifier diodes to generate 48V DC directly -- in both cases this has higher cost and lower efficiency than an AC generator.

Edited October 18, 2022 by IanD

[cuthound]

19 hours ago, IanD said:

 

If you have an AC generator the normal way to connect this to the batteries is via something like a Victron Multiplus/Quattro, which acts as a battery charger and an inverter. This also has the benefit that if the inverter dies you've still got a mains source (direct from generator). And as I'm sure you've found out it's *far* easier to get 230Vac generators (massive market, many models from many suppliers including cheaper ones) then 48V DC ones (tiny market, few models from few suppliers, none cheap). But then a Multiplus II 48/10000 (which you need to charge batteries at 7kW) is advertised at £3400, assuming this includes the latest Victron price rises (which it may not do), so again not cheap...

 

Non-marinised generators (e.g. ex-hospital or standby) tend to use radiators and fans (and a *lot* of air) to keep them cool, which is not convenient on a boat where you really want water cooling --- and for the canals, keel cooling not freshwater. It may be possible to convert them with some difficulty, or may not be. Most marine generators are freshwater-cooled (intended for sea use with a pumped water inlet/output), only a few (e.g. Beta Marine) are available with keel cooling as standard -- though some (e.g. Fischer-Panda, Kohler) have kits to do this buried away in the manufacturers parts list.

 

I agree with the first paragraph. However most (all?) relatively small generators, marine and otherwise are based on plant or automotive diesel engines which can be relatively easily converted to keel cooling.

[IanD]

38 minutes ago, cuthound said:

 

I agree with the first paragraph. However most (all?) relatively small generators, marine and otherwise are based on plant or automotive diesel engines which can be relatively easily converted to keel cooling.

 

Which is what I said -- but unless approved by the generator manufacturer this will void any warranty, which can be an issue with a pricy generator bought new. The same applies with the exhaust, many marine ones intended for sea/freshwater use have wet exhausts, and converting them to dry exhaust for a canal boat can have the same problem -- I know, I asked the manufacturers... 😞

 

For a cheap and/or secondhand one where you want to DIY the conversion, not an issue 🙂

Edited October 18, 2022 by IanD

[cuthound]

1 minute ago, IanD said:

 

Which is what I said -- but unless approved by the generator manufacturer this will void any warranty, which can be an issue with a pricy generator bought new -- I know, I asked.

 

For a cheap and/or secondhand one where you want to DIY the conversion, not an issue 🙂

 

Which is why I suggested to OP looked at ex-BT gensets. Cheap, reasonably well maintained and usually very low hours (typically 12 per annum).

[IanD]

Just now, cuthound said:

 

Which is why I suggested to OP looked at ex-BT gensets. Cheap, reasonably well maintained and usually very low hours (typically 12 per annum).

But as he said, usually too big for him...

[cuthound]

1 minute ago, IanD said:

But as he said, usually too big for him...

The smallest (and most common are in the 10-25kVA range. Most telephone exchanges (which are rapidly being lost due to VOIP) are small rural ones.

 

I agree larger ones (up to 3MVA) and often based on ship propulsion diesel engines would be to large as they are bigger than even a full length narrowboat.