another long post
Not to rain on the parade but working through the physics of it on the flipside when the load is applied and it does speed down then the same amount of power with a larger flywheel will take longer to recover speed.
Sure - no free lunches. F always equals MA, which means that if you needed 10 KW-seconds to start your well pump, an MEP-002A will be running at full output for twenty seconds to come back up to speed after expending that energy. The good news is that the actual motor start demand for something like a two horsepower well pump is more like 10 KW for 100 milliseconds, which would be 1 KW-second, or full power for 1/2 second to recover.
The point is that you've taken a 100 millisecond load (locked rotor amperage on the motor) and spread it out over a second or two. That gives the governor a lot more time to respond and reduces the total amount of droop you're going to get.
In essence the flywheel will only increase the time response it will not substantially change the lowest rpm reached during recovery.
Um, no. If the instantaneous torque required to put x amount of current through the genhead is greater than the amount being made by the engine at that instant, then RPM drops. How far it drops is determined by a) the response time of the governor, b) the amount of time of the load, and c) the amount of instantaneous torque the engine can make, assuming the governor has noticed what is going on and is calling for maximum power. If we're talking about an MEP-002A engine, it probably makes about 16 HP at full power. If you need 17 HP or more for even a millisecond, the RPM will drop over that time span even if the engine is already making maximum power, which it almost certainly isn't.
The governor action, if mechanical, is nearly instant because the mechanical linkage does not flex or stretch much (and we do not have smoke limiters, except turbo engines).
'nearly' instant in a mechanical feedback control system is usually on the order of 50-100 milliseconds. Given that there are 3 power pulses on an MEP-002A per 100 milliseconds, you don't have a lot of opportunities to make more torque during a 0.1 second event unless your governor is really responsive. For the record: highly responsive mechanical governors tend to have bad oscillation and hunting problems. That's why there was a fancy high-tech version of several of the MEP generators that included electrical or electro-hydraulic governors for applications that needed high frequency stability (reduced surge and droop).
The amount of fuel given to respond is proportional to the amount of speed drop. The droop setting can make this more or less. But some is always needed to ensure stability at steady speed. A trade off.
All true. But you're really talking about two different things - the droop adjustment is used to minimize the amount of frequency (RPM) change from 0% to 100% load. Set it too sensitive and you get hunting. We're not talking about droop under continuous load, though, we're talking about time of response to a sudden load change. That's mostly governed by the friction, slop, and inertia of the linkage itself relative to the amount of energy stored in the governor spring, although the droop adjustment does have some effect on that as well. If it helps, consider that reducing the inertia of the governor linkage should increase the frequency of governor hunting (oscillation around the control value) when the droop adjustment is set sensitive enough that it allows hunting to occur.
So in this sense the large engine on a small gen head would be much more responsive and have less droop when hit with a step load. Oversizing the engine is a classic solution to get tighter frequency control. In other words the large engine will hold rpm very well when hit with a big load. Not because of flywheel but because of ability to add much more fuel. Figure it this way: A 100% step load would require 100% fuel. Now double the engine with the same gen head and that same load is only 50% of the engine, yet the engine can respond with 100% fuel or twice the fuel as the small engine.
Again,all true. Basically, you're making the engine capable of making more instantaneous torque with the same injector linkage position change- changing the slope of the governor response curve. If I'm humping five gallon cans of diesel fuel up the hill like I did in the snowpocalypse a couple years ago I think I'll skip the extra fuel consumption at no load, thanks.
As far as witnessing the effect of extra mass just visit youtube and watch all the listeroid and lister videos on those 750 lb, 650 rpm 1400cc 6 hp engines and their little 3kw generators. Their response to load is pitifully slow and their drop in rpm is very noticable. Search Listeroid on youtube.
I don't have to - I own one (a Metro clone of a 6-1 Lister CS). For the record: everybody with the Indian clones has to rework the governor linkage because of sticking and droop problems Dry lubed Heim joints are your friend. Real Listers are better just because the governor was properly set up to begin with, but Lister didn't sell or recommend this engine as a generator engine - they modified it into the
Start-O-Matic, which was the UK version of a
Delco-Light, and included
much heavier flywheels (350 lb each versus about 100 lb each on my clone) on the engine as well as a massive generator pulley as part of the package. Inertia is pretty much the only way to beat the fact that you have five power pulses per second, more or less.
I would think that a quick recovery is as important or more so than the actual surge induced drop in voltage or frequency. Most equipment can handle a big drop if the duration is very short.
Well, maybe yes, and maybe no. My (fancy, new) heat pump complains if things go too far out of spec, flags the power as unstable, and won't start the compressor until it has been stable for a sufficiently long time (minutes). My battery-inverter UPS systems also complain vociferously at motor-start-induced frequency and/or voltage sags.
Another factor is the rpm of the unit. Although low speed is preferred for many reasons it does not lend itself to fast response. 3600 rom machines are inherently faster to respond than 1800rpm or those slow 650 Listers which are pitiful, but love them and would die for one anyway!
They're pitiful as delivered - the Indians use them for irrigation pumps, and really don't care if they're 20 RPM fast or slow at any given time. When you optimize them as gensets they really do a very good job - we have people who've put many thousands of hours on one as a baseline load machine and yes, they do wear out, but they're very cheap and easy to fix. And because they're water cooled, they're perfect for heating domestic water, too.
The big gotcha is that, at the end of the day, a 6/1 really makes only six horsepower at 650 RPM. It can fake you into believing that it can do a lot more because of the huge flywheels (A 6/1 with a 5 KW ST genhead will happily start a 5 HP air compressor), but continuous load had better not exceed 3 1/2 KW or you get smoke and sag for your trouble.
The neat things about a CS (or clone) are that they're quiet (well, easily made quiet anyway), pretty efficient (evacuated crankcase, so the FHP is pretty good), and omnivorous. They'll eat anything that a multifuel will. I want to put ceramic coating on the piston top, chamber top, and precombustion chamber and see if I can get the BSFC down a little more.
Good discussion keep it roling..
Agreed. Sorry for the delay - this fell off my watch list somehow.
steelypip,
So, I could do a round lead mold and fasten it balanced within the fan and have the same effect as hanging another flywheel, without affecting clearance between fan and shroud?
I'd be worried about restricting cooling air flow. Also remember that energy storage is proportional to distance from the crank, so on the inside of the flywheel where the magnet ring is might work. Of course, there's the concern about structural overload, etc to consider.
[/QUOTE]
