My Lister running in full CHP with video - cross posted from VOD

[julesandtash]

The poly vee belt works fine for me - running on the flat flywheel and a power steering pulley on the alternator

Interestingly I found an article comparing the output of various backup power systems .
This image was the waveform of a cheap generator - bears a striking resemblance to my image
Theirs


Mine


whereas this is the output of an APC smart UPS running on batteries


I'm definitely liking the idea of an online (ie dual conversion) UPS if it can rectify the poor quality AC and invert to produce a waveform like that last one

[Dickjotec]

chug
i have a 3kw alternator sitting under the coach somewhere you can have. it needs a frame for the the  drive end as it has a taper shaft to fit to a engine. i have to get rid of the coach so when the scrap man starts on it it should reappear. i have never used it but it was given to me years ago. any use to you?

I modded the front of one of those. A steel plate with a Plummer block bolted on. Worked fine. It has now gone to a new home.

Julianf
I run the 2.5kW alternator on mine with a single V belt that runs on the flat flywheel. I have had no slippage problems. When it was used as a starter belt on the old lister it did slip and I had to use a polly V . I made the pulley for the alt from a aluminium blank.

Dick,

I was going by, when i saw your 6hp, you had (IIRC) a poly-v running arround the outside of the flywheel, with some sort of gel compound on there?

I figured that, if i used a standard auto poly-v, then i could use standard auto pulleys / idlers, and also the belts themselves would probbaly be cheaper / easier to find than others?

That's the angle i was coming from, but id go with whatever you advise?

You are quite correct on the polly v on the 6hp. I used that as it was also from the starter motor, a transit one modified, and I found a V belt slipped. I used belt dressing to increase the friction. It really depends on the speeds and power you need to transmit. For grid tie I need 300+ V and the frequency is not important as it is rectified. I only take 2kW from the pma so I find with the new setup a v belt is fine. Regarding cost there is little difference. If the pulley diameter is small then a polly V is better with a jockey to get more circumference contact with the belt. I would, however, suggest you go with whatever pulley you can source that is the correct size for the rpm you need. I used a jockey on the 6hp as the pma was fixed but with a som base you can tension the belt by moving the alt. on the new setup the pma is mounted with some flex to keep the belt under tension.

[photoman290]

jules,
 if you want to go down the heavy metal route and try a 1 to 1 transformer i have a 3 phase 10 kw auto transformer.
 if you took  the  coil off the centre leg and just used the 2 outside coils you could add a winding to the centre leg and use it as a AVR.
you apply DC to the centre winding to vary the flux intensity, a bit like a adjustable choke, instead of adjusting  the  air gap you adjust the  flux electronically. the DC is derived from the output signal using a dc amplifier. before  big IGBTs came along AVRs used to work like this. i had one that used the idea. you need someone who knows about this stuff. i know a bit but not enough. i do know someone i can ask if you like. just another possible option. i was going to do this myself but apart from the welder my genny output is fine for all my stuff.

[julesandtash]

Possibly Bob, even just as a 1:1 transformer without messing about with AVR windings

I have my eyes on a scope at the moment which will be a useful addition if it goes cheaply

[photoman290]

found the thread i was looking for.
http://www.fieldlines.com/index.php?topic=147181.0
a
the things i was thinking about are called magnetic amplifiers. i will hang on to the auto transformer then,was going to give it to the scrapman. will bring it and the UPS over next time i am up there. can stick them in the van as ballast. need some weight in the back anyway. hit the brakes hard and the back end slides all over the place.

[julesandtash]

That is very kind - you can have some biodiesel in there as ballast on the way home

Next time you come hopefully all will be much calmer (ie not building a playhouse) so we can have a play with the UPS, transformer and the like.
Your AC understanding is far greater than mine.

[julesandtash]

I did some more scoping of the generator today with JulianF's help

In all of these cases, the generator is not connected to the house. The scope is running from the mains supply (not generator) and any loads mentioned are running from the generator

Here is the genuine mains waveform


Here is generator output with no load at all. Nothing like a sine wave at all and more like a mountain range.


Here is the generator output with a purely resistive load (2.2KW electric kettle) only and nothing else. I expected this to be the best waveform but it is really spikey and triangular


and here it is driving a large motor load (2.5HP compressor on load). Despite being inductive, this is the closest to a sine wave and is actually not a bad waveform at all

[therecklessengineer]

I must confess, I don't fully understand exactly how the capacitive excitation works. I found really very little information on it, and only really came away with a vague idea of how it operates.

However, what I can see from the below is that the generator inductance and capacitance are 'ringing' - that's what's giving the lumpy waveform. From what I learnt when I researched it, they're supposed to ring somewhere near the excitation frequency - i.e. 50Hz. It's clearly quite a long way above that.

You could try quadrupling the value of C and see what happens. Check the voltage with a new capacitor though before powering anything from it. My engineering reasoning suggests that will lower the frequency of the ring and reduce the spikes (f0=1/2pi(sqrt(LC))). Although as I don't really fully understand it, I might be wrong!

[julesandtash]

Hi James
Are you sure it should resonate at the fundamental frequency (ie 50Hz)? Some of what I have read suggests the excitation coil should resonate at the third harmonic (ie 150Hz). If so then maybe I need to increase capacitance but not by as much as 4 times.
Motor run caps are not that cheap so I dont exactly want to buy a box full but, as I have a 40uF in there at the moment, maybe I should try a 20uF and an 80uF and see which way improves the situation.

[julesandtash]

I have just ordered 20uF and 80uF motor run caps (1/2 and double the 40uF currently connected and enough, along with the current 40uF to allow me to try 20, 40. 60, 80, 100, 120 and 140 uF if needed by using various caps in parallel) and am bidding on a scope local to me.
It will be interesting to see what effect changing capacitance has and whether it improves things significantly

Hopefully more waveform pictures coming soon

[therecklessengineer]

Actually, that sounds familiar. It's all a bit fuzzy from when I was researching it, and as I said, I didn't really understand it to begin with.

Watching with interest!

[julesandtash]

The caps should be here midweek although we are away at JulianF's next weekend so I may, or may not get some waveform pics up later this week.
If not then should be soon after

[julianf]

The caps should be here midweek although we are away at JulianF's next weekend so...

Have spent most of the weekend plastering, and still the room youre sleeping in has bare stone showing!

...still,we have 5 days yet.

[julesandtash]

Well it looks like James was right with his chosen 40uF excitation cap
I tried 80uF which actually made the wave a bit worse at no load (much greater oscillation at the peaks) and pushed the voltage up to 260V

Dropping to 20uF caused loss of excitation and no output

Looks like a UPS will definitely be the way forward.

Having one a digital RCD tester on eBay for not much money, I will be able to confirm correct operation of all protective devices once I get the UPS installed

[therecklessengineer]

Wow, first time for everything!