Low Output Problem with an MHP

[Hugo Bostock]

I'm looking for advice from someone familiar with MHPs and or generators, the problem is as follows:

We have a micro-hydro plant serving a small fishing community in the Philippines. The water source is around 200l/s, with a net head of around 28m. The turbine is a SKAT T12 crossflow, with an efficiency I would conservatively estimate at 65%. The turbine drives a synchronous marelli generator via v-belt pulleys.

The shaft power at the turbine should be around 28*9.8*0.2*0.65 - 35.7kW

Throw in a loss in the transmission of 5%, gives 32.1kW

The generator is 60% oversized, as recommended by the 'bible' of MHP, Adam Harvey et als Micro Hydro design manual, and is rated at 40kW/52kVA. The community load is pretty inductive with a lot of ice making refrigerators and CFL lighting, probably an overall power factor of around 0.5.

For those who don't know generators are routinely oversized for MHPs because of run-away situations and electric noise from triac switched load-controllers (which this system has).

The problem? The system is only producing 14kVA... And I don't know why! The turbine is professionally assembled and appears to be running well. The water supply is adequate. There is a penstock leak, currently under repair, but the pressure gauge at penstock base reads 50psi, or 34m static head (I don't entirely trust it).

Before we start pulling bits out I would very much like to confirm the generator sizing. Marelli hasn't replied to any of my emails, so I'm throwing it open... What efficiency would one expect from a generator running at 50% load, with a very low power factor? The only set of efficiency curves I've seen suggests the effect of power factor reduces for low loading, but I'd like to confirm it. Any comment on the 60% oversizing? It seems quite high to me... and if I was designing the system I might do it differently as the system experiences low flow during the dry season.  

Any further ideas or suggestions on what might the problem be?

Thanks

[willib]

You havnt said whether it ever worked properly?

[Flux]

Oversizing the alternator should not be a problem and seems desirable with low power factor. In general large alternators are more efficient than small ones and that will partly offset the fact that you are running it at part load.

What have you run out of?  If your speed is falling then you are short of kW, ie you have run out of power from the turbine.

If volts are falling and turbine speed holds up then you have run out of kVA on the alternator due to poor power factor.

You could probably get a large improvement with power factor correction, most alternators are intended to run near 0.8 pf. The 14 figure you mention, is that true kW or is it kVA?

First stage check the speed of the turbine to make sure you have enough power coming from it.

Flux

[Hugo Bostock]

Thanks flux,

OK the turbine is running at the right rpm. The frequency and volts are fine... Just not enough amps going into the load controller. All phases are balanced coming out of the AVR. It looks and feels like a 30kW system... Its just producing less than half that. I have no power factor measurement, so can only give you kVA. I have no  dynamometer, so can't measure the turbine shaft power. The pressure guage on the penstock suggests the pressures ok... but the gauge is definately overreading slightly, so I guess its possibly questionable.

I would have liked to take the top off the turbine to check the runner, as apparently several pythons have been through it, but we were out of sealant and it leaks like crazy otherwise (yes there is a trash screen, but the I suspect the operators have turned it into a snake trap, as the turbine does such a neat job of turning 12ft reticulated pythons into bite-size snake mcnuggets). It's making all the right noises, and the bearings are running at normal temp. The operators swore blind it hasn't been damaged.

Now I think probably some or all of the shortfall is the fault of the turbine, as the generator appears to be running smoothly (I can't think of a problem with the generator that would show as an evenly balanced reduced power output...) What I don't know is what kind of generator efficiency I should be expecting... I thought a quality generator was good for 75% at 25% load and 0.8pf, so would have hoped for an efficiency of around 60%+? (I have no basis whatsover for that guess) that would mean the pulleys delivering 14.5kVA/60% = 24.2kW, or a turbine shaft power of at least 25.5kW, or a turbine efficiency of:

(actual power: 25.5kW) / (available power: 200 (l/s) * 30 (m) * 9.81)

or 43%... Which is dismal! On the other hand, if the turbine is running at 75% efficiency, then the generator is running at around 14.5/(58.9*0.95), or 26% efficiency - which is equally depressing... If its the turbine, then I guess it must be damaged and will need to be remachined. If its the generator, then its probably just a characteristic and we'll have to consider a less oversized unit and maybe an expensive run-away prevention system.

So if I understand you right, if the generator is working too hard against the power factor, that'll show as a voltage drop?

The MHP is at a pretty inaccessible site so next time I visit I want a pretty good idea of what, if any, remedial work is needed before I depart.

Thanks for your help

[Hugo Bostock]

Haven't I? Sorry, yup to all intents and purposes it appears to be working fine... No horrid noises, no red-hot parts or stalling... just too few amps!

[Flux]

Hugo

I don't follow this. You say volts and frequency ( speed ) are ok. Is that when running normally or when you are short of power.

If the turbine is holding the alternator nominal speed under all conditions it is doing its job and you can't ask more of it.

If the alternator is maintaining nominal volts under all conditions, again it is doing its job. If the volts collapse then it isn't.

Current is a secondary issue, if the alternator volts never drop then lack of current must be an issue with the load or the load control. The alternator will continue to supply increasing amps up to the point where its volts start to fall.

For short periods it will supply well beyond nominal amps.

If you are right about frequency and volts then I think you may have a load controller problem, but even then it will show as low volts on the load.

Flux

[drdongle]

You still haven't said wither the system has always delivered the output you describe or at one time put out more power. If it has always delivered the same power then it sounds like some one goofed on a calculation of what the stream could deliver.

[willib]

Thats the problem , Hugo you have snake head bits in your turbine , sorry couldnt resist..

Seriously though , you have no data to go on , thats a problem.

You havnt inspected the impeller yet.

I suspect the impeller is chewed up .

From what you said that has gone through it, i'm surprized it works at all..

if thats not the problem , and it turns out that the impeller is fine, Triacs have been known to fail, as any semiconductor can .

Later on ,when you get it going maybe you could try adjusting the power factor to correct for the predominently inductive load.

[Gary D]

Fellows, perhaps the generator is working fine? Hugo hasn't said if the volts at the end use are low... perhaps the transmition line calculation was done only one way, not both. Line losses could have the load dump thinking usage was small and prematurely dumping needed power if this is the case? Just a simple voltage reading while power is being consumed (end use area)would disprove my thoughts. Maybe not worth mentioning... simple thought from a simple mind here... Gary D.

[Nando]

Hugo:

Connect directly with me and supply, as much info as, you can.

I do not want to guess what is going on, let's analyze the problem.

My address is on the message heading

Regards

Nando