home made 12 volt generator

<< < (9/11) > >>

captain ron:
 Quote from: Homegrowndiesel on January 27, 2007, 08:06:19 PM

Hey CR,

The only reason my Homemade DC Generator is loud is because it is on display, without any shielding, or sound control. ;D

 Yea I like the vrod approach. I thought a Harley engine would draw some attention. Liquid cooled is a must. Easier to recapture the thermal energy also.

With the 5, or 6 speed trans you could easily choose the ratio to efficiently drive air conditioning, electrical generation, and an air compressor.

I would be willing to help.


It may take a while to find an affordable V-Rod engine and transmission as they are still too new not a lot of salvage. but I will find one and build this. It will be fun, cool, different and a great conversation piece. And when I find all the HD components I will call on you for your offered help. I keep watching E-Bay and Crashed toys .com and will put a bug in my friends ear that owns the HD dealership

As per my earlier post on this thread, I very much like the idea of a simple 12v / 24v DC generator for use as a supplementary battery charger; personally it would only be for use as an emergency back-up as I already have a full-size silent generator for 240v mains voltage (and also, as it happens 12v DC, but only at 8 amps).

I've been looking through the catalogue of my usual supplier of workshop machinery, and I see that 4-5hp Honda petrol engines are far cheaper than I would have expected, and they are ready-to-go with pull-start, fuel tank, varous drive pulleys etc. I realise these will be noisy and shakey little things, but for my application (ie. emergency use) they would be ideal. Carrying petrol isn't really a problem for me as my full-size generator is petrol (yes, I would have preferred a diesel genny, but petrol option was much smaller, quieter, lighter and cheaper for the same output).

As well as various low and high voltage alternators, the same catalogue also includes lots of other devices that the engines can be bolted to, such as air compressors and various types of water pump, including jet-washer pumps - it's a shame I cannot really think of a good use for any of those things on the bus.

I still think my earlier idea of using the back-up engine to somehow drive the OEM bus alternator would be the perfect solution, but I also like the idea of using two small and cheap 12v car alternators rather than a heavy and (probably) expensive 24v one. My house batteries are 12v, so there would be benefits there too.

I couldn't help but wonder why I hadn't come across petrol-powered battery chargers / DC generrators before, given that they seem such good idea. I did a quick search on Google, and the first result that came back was of this unit from Australia:

In order to choose an engine to power such a generator / charger I would be very grateful if someone could offer a clear explaination of how to calculate the engine horsepower required to turn an alternator (or alternators) of a given output. Presumably, if the efficiency of the alternator is known it should simply be a matter of converting mechanical power to electrical power? Also, what is the optimum rpm for a car-type alternator? I've always assumed their power output was very flat over a wide range of engine rpms, but presumably there is a speed they are typically most efficient at. I can work our the gearing required between the engine and alternator myself.

Lastly, I very much like the idea of the engine throttle being automatically opened and closed as the load on the alternator varies - in fact my full-size generator has this feature and it is a major benefit. Can anyone think of a simple way to acheive this? Designing custom circuitary to operate servos or stepper motors is beyond me.


Tim Strommen:
Shoot - I just spent 30 minutes writing up a big long explanation of how this is done, and then pushed some funky keyboard combination and lost the whole post by going "Back" (GGGGRRRRRRR  >:().

Oh well (the milk is spilled) - I'll try a truncated version.  You'll want to have these four equations handy:

Watts = Volts X Amps
Drive Ratio = Alternator RPM/Engine RPM
1KW (1,000 Watts) = 1.34102209 HP
Horsepower = Torque x RPM / 5252

There is also a rule of thumb you’ll want to remember: Assume a 20-25% loss at every conversion.  This would be mechanical-to-electrical like in an alternator – and even mechanical to mechanical like in a belt drive (some belt drives can be up to 98% efficient, but if they are not maintained can drop to 92% or less efficient – in my opinion I’d play this one safe).

You’ll want to start at the alternator as it’s the last conversion (I’m disregarding wire “line” loss for this discussion as it should be part of the system design equations already :) ).  Make sure the alternator you pick is sized for the system and charge-rate of your system.

For this example – I’ll assume we’re using a 24Volt (28 volt charging voltage) 200Amp alternator which is belt driven.  Also – I’ll assume that the max current is achieved at 6000RPM (this information would be in the alternator’s spec-sheet).  This means that the alternator will create 28Volts x 200Amps = 5.6KW at the output posts.  However this is a result of a mechanical input shaft to electrical field conversion.  Which means – you’ll be drawing 5.6KW x 120% (added 20% of loss to the output power – or a decimal value of 1.2) = 6.72KW at the pulley.  If you don’t have an engine that’ll put out the required HP at 6000 RPM ( or you don't like the "sound" {punn} of an engine running at 6K RPM for any length of time) – then you’ll need to use a step-up drive system (belts are typical).  So let’s add another “loss” to the system: 6.72KW x 1.2 = 8.064KW.  This number will let you do two things: 1) pick an appropriate peak output engine, and 2) a suitable drive system.

To pick the engine, convert the KW to HP (KW x 1.34102209 = HP) so: 8.064KW x 1.34102209 = 10.814HP.  You’ll need to pick the next size UP from this or you’ll stall out the engine at max load.  Let’s say that we have selected an 11HP engine.

To pick the drive system – you need to figure out the max power and RPM of the engine from the published performance table.  Now figure out the pulley size by measuring the alternator pulley size and getting the circumference (let’s say the alternator has a 2” diameter pulley).  The circumference of our alternator pulley is thus: Pi x Diameter = Circumference or Pi x 2” = 6.283” circumference.  Now figure out your required drive ratio by finding the RPM at which your engine creates just a bit more than the peak power required (we’ll say 11HP in this case – and for this example say that the hypothetical engine's performace table indicates that this occurs at 2600RPM).  So your peak output of the alternator (28Volts@200Amps) occurs at 6000RPM on the alternator and the 11HP is generated at 2600RPM.  So find your drive ratio by 6000 / 2600 = 2.308:1 (or 2.308 rotations of the alternator for every 1 rotation of the engine).  To find the size of your engine (drive) pulley – take your drive ratio number (2.308) and multiply it by the alternator pulley circumference (6.283”): 2.308 x 6.283” = 14.501” engine pulley circumference.  You can get your pulley diameter now by reversing the math: 14.501" / Pi = 4.616” diameter.  You’ll also need to ensure that your drive system can support the power rating for the length of time it’s being used (i.e. 11HP under continuous use).  Some systems will require the use of a dual slot pulley (or worse if really big).  You also need to ensure your engine can support the side-load of the tension of the pulley/belts against the crankshaft bearings (or the engine will fail pre-maturely).  You may need to use a short shaft and a set of pillow blocks (aka bearing blocks) to take up the belt load from the engine.  Just be sure to align your pulleys so they don't eat belts quickly.

Both your selected engine and selected alternator should come with performance tables showing (for the engine) how many HP are generated at a certain RPM – and (for the alternator) how many Amps are generated (at the charging voltage) per input RPM.  You’ll need to match up values of the alternator to the values of the engine.  This is easily done within a look-up table in a microprocessor (and you'll probably need one to controll this well - as most of-the shelf voltage regulators are not suitable for this task).  You’ll also need to get the table for the field current of the alternator to the output amps.  Again – plugging these into a microprocessor’s look-up table is easy.  A voltage sensor can tell the microprocessor if the voltage is below the nominal rating (less than 28V because of draw-down) and increase the RPMs of the engine.  A smart-desiner's voltage regulator can also prevent the field from getting a bunch of current until the engine is at an RPM where it can support the requested current (a fractional multiplier equation [to derive a "margin"] will ensure that the engine power [throttle] is always that fraction above the actual requested alternator load - and then only allowing the requested output to be so-output in relation to the >>actual<< engine RPM via the look-up table will prevent a stall).

By adjusting the multiplier (engine HP to Alternator output "margin") you can affect the rate at which the engine will obtain it's desired RPMs.  If you want to get more complicated - you can vary the magnitude of the multiplier over the base value (with another multiplier) by evaluating the difference of the requested current in relation to the actual output current.  This way it will push the engine up faster if a large load is detected - but will only rev up the engine slowly under light loads (rate of acceleration affects fuel economy - and if it's a light load, it's more likely it won't negatively affect the battery pack's charge state with a huge draw).  This would probably be optimal, as allowing the batteries to take care of long term variations in smaller amplitudes will stabilize the system faster.

This system can be further improved with load current sensors (current loops or a shunt) to tell the microprocessor exactly how much power is being drawn from the alternator (and even another to show the ammount and direction of charge/discharge from the battery) – which will give it a better picture of the power situation in-circuit.  For example – say you turned off your 2400Watt inverter and now there is no draw on the alternator other than the trickle the batteries need – the shunt will note this change quickly and the voltage regulator can drastically cut the field current to the alternator to prevent a voltage spike, keeping only the trickle charge to the batteries.  Also battery and ambient temperature sensors can be used to better match charging rates to the current battery states (and with a microprocessor controlling the alternator - you can do those fancy multi-stage charging steps).

If you have the microprocessor running the engine and watching the system voltage/current already - you may as well put in some other features like engine warm-up before loading, engine cool-down before shut-down (required for turbo longevity if one is used on your engine), low-oil shut-down, over-temp shut-down, low-fuel shut-down (if the generator and Prime-mover are run from the same tank), auto-start, auto battery-maintain, "quiet time" (hook up a clock and set an "out of bounds" time-frame when most in your camp ground are sleeping...), battery fan controls, etc.

Of course building your own voltage regulator is a bit complicated, and making a mistake, and/or not having hard fail-safes (fuses, voltage limiters, diodes, etc.) can destroy very expensive parts in short order (say a $3000 inverter? or a large battery bank?).

It’s not for the faint of heart – but it can still be done with a good deal of planning.

Heck – diesel locomotives use a similar control system to regulate the speed of the engine/generators – and Honda has produced small A/C inverter-generators that do this all the time (so it has been done, and can be done again).

Do it your way – and be safe (put in emergency shut-downs and fire/smoke detectors), there's nothing worse than burning down your bus with your family in it…  :'(

Oh - almost forgot.  If you use a current shunt on the output, it creates a voltge across it.  If you tap in a low-voltage solenoid/actuator to the measurement lugs of the shunt - the voltage could theoretically push the solenoid out harder/farther with a higher load.  Attaching the output shaft of the solenoid to the throttle is the easiest way I can think to automatically control the speed of the engine based on the load - but this would have to be coupled with an anti-lug govenor (which I think most small engines have in them anyway).  Sounds complicated and hard to ballance - mostly because is is :).  Manufacturers who use this method have custom actuators and throttle springs for this function.

Another concept that I've not seen but is concievable - is to drive and actuator from the voltage regulator's field output (more field current - equals more alternator load, and should be matched with more throttle movement) - this could possibly stabilize a generation system faster (but don't quote me on this :-*).



P.S. Hope this wasn't too long-winded and boring... :)

^^^^^^^ ;D ^^^^^^ That's what I'm talking about.^^^^^^^ ;D^^^^^^^
You sir, have placed in a "nutshell" the most important "basic" factors to consider needed to plan such an interesting project (though not for the faint of heart) without the "confusing micro-details" that would have earned me a PhD in a few more weeks. The exact calculations and formulas that you presented to derive at correct equipment combinations is the first in this board, bar none.

TIM STROMMEN, You are an asset to this community. The substance of your post exceeded all expectations in this thread. I salute you!

Tim Strommen:
Hey Gr8njt,

    Thanks for the note :o, it was rather entertaining to read (and all this after the thread almost got away from Captain Ron).

The accidentally deleted post I wrote previously may have given you the extra edge to get that PhD  ;).




[0] Message Index

[#] Next page

[*] Previous page