Alternative Power sources - Where does "surge" capacity in Generators come from?

Author

Where does "surge" capacity in Generators come from?

philkryder

2005-12-15, 10:21 pm

I've seen several posts that relate the ability of a generator to
handle a surge load to the stored rotational energy in the engine and
generator itself.

This has caused me to wonder about a few things.

How important is the stored rotational engergy compared to:
1) governor responsiveness
2) Torque rise of the engine.
3) Generator design

Specifically, how much energy IS stored in a typical generator running
at 1800 rpm versus how much is stored in a lighter generator running at
3600 rpm?

Does the energy stored go up with the square of the rotational speed?
Does the enegy stored go up linearly with the mass of the rotating
components?

Also, how important is constant voltage - versus constant cycles when
trying to start a reactive load?

Thanks
Phil

RF Dude

2005-12-16, 12:21 am

I think you are drawing some conclusions based on what inverters usually
market as "surge" capacity and 5-minute capacity. Inverter 5-minute
capacity is based on heat inertia, and its "surge" capacity is only
milliseconds as the large capacitors in the high voltage DC section
discharge to supply a momentarily overload.

For generators, I can think of three ratings: Prime (what can be supplied
indefinitely), Standby (what can be supplied for 1-hour), and motor
starting/fault clearing overloads. You can imagine that the 1-hour rating
is a thermal limit. Someone else may have a comment here about what I'm
about to say, but it appears that the 10% rating puts you closer to the full
rating of the conductors and protective devices. That is, prime rating
would represent 80% of the circuit breaker rating, while the standby rating
would be closer to the service rating of the wire and CB which is acceptable
for short durations only. These ratings are affected by the engine and
cooling system, as well as the thermal characteristics of the alternator
winding... for instance, do you want 105C rise, or 135C rise, type of
insulation, etc.

Regarding reactive loads, you can also add an optional Permanent Magnet
Generator (exciter) to the end of the alternator that will ensure you up to
300% rated current for a few seconds to start heavy inductive loads or clear
a fault. The permanent magnets in the exciter will keep the field from
collapsing, and thus, loss of output.

You have to consider the type of load, crest factor and harmonics when
sizing a generator. Remember that the engine supplies POWER or kW. VARs
are supplied by the alternator. A large amount of current to a poor PF load
doesn't necessarily tax the engine. So it isn't uncommon to put a 50 kVA
alternator on a 30 kW engine if the VAR's are required. Another reason for
oversizing the alternator is to keep the voltage THD under control with
non-linear loads. The alternator and wiring must be rated to supply the
current. The engine must be rated to supply the power.

Then you have manufacturer engine ranges. For instance, a Wilsons generator
with a Caterpillar 4 cyl diesel, no turbocharger will have the same engine
in a 30kW or 40kW generator. Only the amount of copper in the alternator
changes. The 36 kW (Prime) 40 kW (standby) engine/alternator combo will
supply 44 kW for a while... make it 45 kW and the engine slows down and
stalls. Go to 50 kW, the SAME engine gets a TURBOCHARGER... go to 60 kW,
the same engine gets an oil cooler. The 50 kW generator with PMG will
supply 60 kW until the CB opens because the engine is strong enough to turn
the alternator.

If you were asking this in relation to small consumer generators, things are
not so controlled. Quality and design take a back seat to affordability.
Anything can happen.

"philkryder" <alt.google@Kryder.com> wrote in message
news:1134699095.236787.259520@z14g2000cwz.googlegroups.com...
> I've seen several posts that relate the ability of a generator to
> handle a surge load to the stored rotational energy in the engine and
> generator itself.
>
> This has caused me to wonder about a few things.
>
> How important is the stored rotational engergy compared to:
> 1) governor responsiveness
> 2) Torque rise of the engine.
> 3) Generator design
>
> Specifically, how much energy IS stored in a typical generator running
> at 1800 rpm versus how much is stored in a lighter generator running at
> 3600 rpm?
>
> Does the energy stored go up with the square of the rotational speed?
> Does the enegy stored go up linearly with the mass of the rotating
> components?
>
> Also, how important is constant voltage - versus constant cycles when
> trying to start a reactive load?
>
> Thanks
> Phil
>

philkryder

2005-12-16, 12:21 am

"I think you are drawing some conclusions based on what inverters
usually
market ..."

Sorry that I left you with that impression.
That was not my intent.
I was trying to ask questions about generators - not voice conclusions.

"....For generators, I can think of three ratings: Prime (what can be
supplied
indefinitely), Standby (what can be supplied for 1-hour), and motor
starting/fault clearing overloads. "

I was trying to refer to the 3rd item that you list. "...motor
starting/fault clearing..."

I've seen folks comment that the rotational mass of the unit affects
the ability to supply surge.

I was skeptical of this because I thought that the only way to "obtain"
that stored energy was for the RPMs to drop - which would affect
cycles.

Thus, I was lead to wondering:
How much rotational energy is stored in the unit?
How much of it can be used to start motors?
and,
Are these quantities significant when compared with other things like:
Normal load,
Torque rise,
Governor responsiveness and
Generator design.

"....Regarding reactive loads, you can also add an optional Permanent
Magnet
Generator (exciter) to the end of the alternator that will ensure you
up to
300% rated current for a few seconds to start heavy inductive loads
...."

This "seems" like it would be a "good thing" - what is the down side if
any?

RF Dude

2005-12-16, 2:21 am

I can't respond directly to the rotational aspects, but I don't think there
is much stored energy in the rotating machinery. At least not much beyond a
few cycles at best. Look at the published specifications. One of them is
the allowable voltage and frequency dip for a "step-response". I would
often just test this by turning on a switch that takes the machine from 0 to
100% load while capturing the sag on datalogging equipment. You can hear
the machine dip and recover speed and it does so very quickly. How much of
a dip you can accept depends on the load. You get a surge when the load is
disconnected depending on how quickly the automatic voltage regulator can
back it off.

Have you every listened to the ice cream truck generator? You hear that dip
whenever the dude starts to pour a cone. Mechanical vs electronic governor,
and turbo lag tend to be the factors.

PMG? Cost. Some manufacturers offer it as an option. Others (like Kohler)
build it into their alternators as standard. For instance, the residential
Kohler (12RES) which is 10.5 kW claim they can start a 4 tonne air
conditioner (probably not much else can be on). Don't know of any other
issues with the PMG. Recall that a compressor is essentially a locked rotor
for a few cycles drawing 120A for a 4 tonne unit. That 120A doesn't have
much to do with power. The compressor is very inductive as it begins to
spin.

m Ransley

2005-12-16, 10:21 am

I cant answer your question but point out on my 7500exl at 7000w for 5
minutes load the exuast was glowing red about 4 inches out the head on a
50f day. After all its air cooled, I wonder what engine temps would be
on a 90f day 90% humidity. And a test on a 4000exl for load and surge
failed on a Consumer Reports test, avalaible online. I personaly view
surge load as something I never want to use as I like to run equipment
conservativly. Ive seen other home gens just not start something near
surge, Im sure it stresses every component. It isnt like buying a
furnace where you have an independant AFUE rating and can be reasonably
confident of performance.

Pop

2005-12-16, 10:21 am

Inline, my best efforts:

"philkryder" <alt.google@Kryder.com> wrote in message
news:1134699095.236787.259520@z14g2000cwz.googlegroups.com...
: I've seen several posts that relate the ability of a generator
to
: handle a surge load to the stored rotational energy in the
engine and
: generator itself.
:
: This has caused me to wonder about a few things.
:
: How important is the stored rotational engergy compared to:
: 1) governor responsiveness
===> The more stored energy, the more accurate the rpm can be
maintained, but once changed, the longer it takes to recover rpm.

: 2) Torque rise of the engine.
===> don't know

: 3) Generator design
===> Lots of variables, I suppose; stored "rotational energy" is
no more than the flywheel effect, so the pros are more constant
rotational speed, but the cons are tougher recovery when
something does change the speed.
:
: Specifically, how much energy IS stored in a typical generator
running
: at 1800 rpm versus how much is stored in a lighter generator
running at
: 3600 rpm?
===> I'd imagine it depends a lot on the designer. If there was
anything inherant about it, I think you'd see it hyped in the
advertising.
:
: Does the energy stored go up with the square of the rotational
speed?
: Does the enegy stored go up linearly with the mass of the
rotating
: components?
:
: Also, how important is constant voltage - versus constant
cycles when
: trying to start a reactive load?
===> The lower the frequency drops, the more current is drawn,
and the harder tha machine has to work, so if you approach the
specced voltage/curent, the importance is going to rise.
I don't think frequency would be tha timportant within
reasonable specs, but of course, again, it creates varying
current draws. The higher the frequency the more efficient (eg
the 400Hz used on aircraft/ships), but during starts I'd think
only the current mattered. Lower frequency, less energy for the
same V.

Or were you looking for more complex answers?

HTH,

Pop

:
: Thanks
: Phil
:

nospam@nouce.bellatlantic.net

2005-12-16, 11:21 am

On 15 Dec 2005 18:11:35 -0800, "philkryder" <alt.google@Kryder.com>
wrote:

>I've seen several posts that relate the ability of a generator to
>handle a surge load to the stored rotational energy in the engine and
>generator itself.
>
>This has caused me to wonder about a few things.
>
>How important is the stored rotational engergy compared to:
>1) governor responsiveness
>2) Torque rise of the engine.
>3) Generator design
>
>Specifically, how much energy IS stored in a typical generator running
>at 1800 rpm versus how much is stored in a lighter generator running at
>3600 rpm?
>
>Does the energy stored go up with the square of the rotational speed?
>Does the enegy stored go up linearly with the mass of the rotating
>components?
>
>Also, how important is constant voltage - versus constant cycles when
>trying to start a reactive load?
>
>Thanks
>Phil

Taking this from a more simplistic point of view. Surge is about
horsepower. You cant get 10000W from a 10HP engine.

We assume the surge rating is below the limits of copper, alternator
and protective devices (fuse/breakers). This is fairly likely for the
under 10kw class units.

What it's really about for the coleman and try built class of low cost
alternators is the amount of sustained load the engine can support
and the peak load the engine can support. Obviously a 5000W
gen with a 7HP engine is running near peak all the time, however
if the engine turning it is 10HP there will be enough power to turn
the alternator even with an 8000W load for short period of time.
Add to that a there is at least some greater amount of increased
mass in larger engines yeilding more inerta to support larger
spikes in load.

This is where inexpensive genset get in trouble. Small engines
are power limited and its easily possible to get a 3000W gen
loaded near peak and be using all of the available horsepower.
At which point any increase in load will cause the engine to slow,
output voltage and frequency to drop and load currents usually
increase under those conditions causing operating conditions
to further decay. If the same genhead were hooked to a greater
horsepower engine and still rated at 3000W then the load increase
can be greater before things reach limit conditions.

For example when I shopped for a emergency power gen
I found many to choose from in the cheap unit bracket but a
Troybuilt 5500W was one of the few that had a bigger 10hp
engine. The 8250W surge claim comes from have the
horsepower to turn the gen head. [please no comments
on good unit or bad. It's only for short duration (hours) use.]
It's performance to date is far better than a friends Coleman
of 5000w rating but far lower horsepower engine. The
difference shows when there is a large load start like motors.

The worst unit I'd ever use was 30 years ago. A 2000W unit
with a 3.5hp Techumsah, at full load the ending was running
with the govenor full open. Typical use on building site was
3-4 60W lights and power tools. When that engine wore out
(didn't last running at or very near full bore all the time) a larger
5HP engine was fitted and that ran for years with no problem
supporting a 2000W load. No attempt was intentionally made
to run greater loads but powering tools like skillsaws and
table saws you saw/heard he difference as you loaded the
unit. With the added horsepower it was possible (untested)
to get more continious power from the gen head but at risk
of it's failure. I do know that when starting motors we had far
more power avalable with the 5HP and less trouble with
blades sticking when cutting.

Allison

nicksanspam@ece.villanova.edu

2005-12-16, 11:21 am

m Ransley <ransley@webtv.net> wrote:

>... I wonder what engine temps would be on a 90f day 90% humidity.

Generators don't sweat :-) Maybe small inverter versions with higher rpms
have higher surge to rated power ratios because they can speed up faster
to meet a surge load.

Nick

daestrom

2005-12-16, 6:21 pm

"philkryder" <alt.google@Kryder.com> wrote in message
news:1134699095.236787.259520@z14g2000cwz.googlegroups.com...
> I've seen several posts that relate the ability of a generator to
> handle a surge load to the stored rotational energy in the engine and
> generator itself.
>
> This has caused me to wonder about a few things.
>
> How important is the stored rotational engergy compared to:
> 1) governor responsiveness
> 2) Torque rise of the engine.
> 3) Generator design
>
> Specifically, how much energy IS stored in a typical generator running
> at 1800 rpm versus how much is stored in a lighter generator running at
> 3600 rpm?
>

The energy stored in a rotating system is 1/2*I*omega^2. The 'I' is moment
of inertia, a measure of the mass of the rotating elements times their
average distance from the center of the rotating shaft (sort of). So for
the same mass, about the same shape, a 3600 RPM machine has four times the
kinetic energy stored in the rotating elements.

> Does the energy stored go up with the square of the rotational speed?
Yep.
http://hyperphysics.phy-astr.gsu.edu/hbase/hframe.html

> Does the enegy stored go up linearly with the mass of the rotating
> components?

Not quite. It goes up linearly with the moment of inertia. That is also
linear with mass *if* the average distance of the mass from the center of
rotation stays the same. Or you can have the same mass, and just sort of
'extend it' outward from the shaft. A 2 lbm shaft has lower moment of
inertia than a 1 lbm shaft with a 1 lbm, thin disk mounted on it. The thin
disk means that some of the mass is further away from the centerline of
rotation. That's why engine flywheels are thin disks.

>
> Also, how important is constant voltage - versus constant cycles when
> trying to start a reactive load?
>

Well, starting a 'reactive' load doesn't impose any sort of delta-energy or
frequency problem. But I suspect you're thinking more of a motor than a
purely 'reactive' load like a transformer or capacitor bank.

There are all sorts of different ways to measure a generator's capabilities,
some of them more obvious than others.

There is a maximum current that the windings can take. You can overload
these a bit for a short time, because it is the overheating of them that
damages them. And in a short duration overload, the temperature doesn't
rise too fast, so there isn't much damage to the insulation of the windings
as long as it's not too long.

There is a limit as to how much the AVR and/or field winding can take. This
limits how much reactive load you can supply (low pf, not necessarily
motors). If the exciter is simple and uses generator output voltage to
supply the field, then a large load connecting all at once will 'dip' the
voltage so low that the exciter can't supply the field and you have a
'voltage collapse' situation. Top dollar units use a current transformer
and additional rectifier on the main line to help feed the field, so a
sudden surge in current supplies a sudden surge in field current to help
sustain the output voltage. This also helps with fault clearing scenario's.
With a simple exciter, the large current of the fault just causes voltage
collapse before the fuse/breaker can trip. Expensive units, the short term
capability of the 'current enhanced' exciter can provide several times the
normal current long enough for protective devices to operate. But even the
simple ones, once voltage collapses, there is little fault current left.
It's just a matter that the expensive ones can clear the fault and power
other remaining loads.

There is also a limit on the 'prime mover', in most cases an engine. After
all, you can't get more out of the generator than you put in, so if the
engine can only supply 5 hp, then you can only get 3730 watts, no matter
what. Putting a small engine on a big gen-head means you're 'wasting
copper', while putting a huge engine behind a small gen-head means you're
'wasting iron'. If it weren't for starting/surges, ideally you could match
the engine to the gen exactly. But to start a large rotational load, you
need some excess 'iron' (i.e. the engine hp capabilities need to be higher).

Starting a large motor isn't exactly the same as a 'large reactive' load. A
choke or cap. will only draw large current for a cycle or two (for details,
study 'transient reactance calculations'). But a motor can draw large
amounts of *power* (not just current) for as long as it takes to accelerate
up to speed. If it's uncoupled from any load, that can be as little as 1-2
seconds. If it's driving some big heavy load, it could be 5-15 seconds.

Now, if the moment of inertia of the load you're starting is small compared
to that of the generator, and the governor is quick-acting, when you start
the load, the energy in the genny's rotor flows through the wires to the
motor and starts accelerating it. This in turn slows the genny, and the
governor quickly increases the engine power to bring the genny back up to
speed. The amount that frequency 'dips' is small, and the time to recover
is short.

If the motor's inertia is large, then the 'dip' in frequency will be more
severe. If the load on the genny is near the max of the engine, the
governor will open the throttle full, but the engine will take some time to
bring things back up to speed. So the frequency disturbance is more severe,
and longer to recover.

One large DG set that we have is rated for 4000 kw, but it is used to start
up *one* load, at 4000 hp (3600 rpm) with a massive rotor.. We start the
engine up and have it up to 900 RPM (60 hz for that unit). When we close in
the load, it starts accelerating the motor, but the DG 'dips' down to about
600 RPM (40 hz). Once the motor is up to about 2400 RPM, the governor
(which goes 'full open') brings the engine and motor up to rated speed over
the next 20 seconds. The engine is actually overloaded as it runs back up
to speed, but it is a short duration and the manufacturer okays this
service. Once up to speed, the governor cuts back on the 'throttle', and
maintains the engine/load pair at the proper speed/frequency. It's a
dramatic example of a load who's moment of inertia and load are very high
compared to the genny.

daestrom

RF Dude

2005-12-16, 9:21 pm

See below:

"daestrom" <daestrom@NO_SPAM_HEREtwcny.rr.com> wrote in message
news:4ZGof.31598$XJ5.20014@twister.nyroc.rr.com...
>
> Well, starting a 'reactive' load doesn't impose any sort of delta-energy
> or frequency problem. But I suspect you're thinking more of a motor than
> a purely 'reactive' load like a transformer or capacitor bank.

A motor that isn't moving (or locked rotor) IS A TRANSFORMER. Primary is
the stator winding, and secondary is the rotor winding. The power factor is
down at 0.2 or even worse.

> Putting a small engine on a big gen-head means you're 'wasting copper',
> while putting a huge engine behind a small gen-head means you're 'wasting
> iron'. If it weren't for starting/surges, ideally you could match the
> engine to the gen exactly. But to start a large rotational load, you need
> some excess 'iron' (i.e. the engine hp capabilities need to be higher).

> Starting a large motor isn't exactly the same as a 'large reactive' load.
> A choke or cap. will only draw large current for a cycle or two (for
> details, study 'transient reactance calculations'). But a motor can draw
> large amounts of *power* (not just current) for as long as it takes to
> accelerate up to speed. If it's uncoupled from any load, that can be as
> little as 1-2 seconds. If it's driving some big heavy load, it could be
> 5-15 seconds.

Your comment conflicts with what I stated in my previous post. Starting a
large motor does require a large alternator (or copper) to supply the VARS
(current) since the motor PF is so low. Once the motor is turning at speed,
it's PF is around 0.8 so then power is dominant. In other words, current
and voltage are severely out of phase with each other during starting. The
power vector is small and gets larger as the motor spins up. Well, this is
a broad statement. It also depends on the type of motor and winding. Also
keep in mind that with high currents, I^2 x R losses are a component that
eats power in both the generator winding and other wiring.

Maybe I'm wrong... I may be twisting my head around this to much from an
electrical rather than mechanical angle. Anyone else care to comment?

> One large DG set that we have is rated for 4000 kw, but it is used to
> start up *one* load, at 4000 hp (3600 rpm) with a massive rotor.. We
> start the engine up and have it up to 900 RPM (60 hz for that unit). When
> we close in the load, it starts accelerating the motor, but the DG 'dips'
> down to about 600 RPM (40 hz). Once the motor is up to about 2400 RPM,
> the governor (which goes 'full open') brings the engine and motor up to
> rated speed over the next 20 seconds. The engine is actually overloaded
> as it runs back up to speed, but it is a short duration and the
> manufacturer okays this service. Once up to speed, the governor cuts back
> on the 'throttle', and maintains the engine/load pair at the proper
> speed/frequency. It's a dramatic example of a load who's moment of
> inertia and load are very high compared to the genny.

This is an interesting example. There is something more going on here.
Like you are using the engine and alternator limitations (impedance) in lieu
of the usual motor starting methods to control starting current. Do you
start this large motor using the public utility grid? It would have to have
a sophisticated control system to prevent overcurrent and unsafe conditions
since the public utility is considered an "infinite bus" (ideally capable of
supplying whatever you ask of it).

RF Dude

daestrom

2005-12-17, 3:21 pm

"RF Dude" <post@thisnewsgroup.com> wrote in message
news:uIIof.5518$PQ3.846981@news20.bellglobal.com...
> See below:
>
> "daestrom" <daestrom@NO_SPAM_HEREtwcny.rr.com> wrote in message
> news:4ZGof.31598$XJ5.20014@twister.nyroc.rr.com...
>
> A motor that isn't moving (or locked rotor) IS A TRANSFORMER. Primary is
> the stator winding, and secondary is the rotor winding. The power factor
> is down at 0.2 or even worse.

Yes, it is a transformer, with the secondary winding short-circuited. Since
the secondary has both inductance and resistance, it's overall power factor
is low, as you state. But an ordinary transformer is either unloaded, or
it's pf is determined by what is connected to it. In the motor's case, the
'load' is the shorted windings on the rotor end bars.

Even with the low pf, if you work out the *power* drawn by the motor when it
is stalled, you will find it is much higher than normal running power.

So motors are a 'double-whammy'. They are a 'reactive load' in the sense
that the secondary has some reactance and starts with a low pf, but they are
also a large 'real' load in the sense that they require a large amount of
watts when starting.

The low pf can cause 'voltage collapse' if the exciter can't handle the
'reactive' part, and they can cause a frequency dip if the
prime-mover/governor can't handle the 'real' part of the load. Of course
the magnitude of the problem is a ratio of the generator size versus the
motor size.

>
>
>
> Your comment conflicts with what I stated in my previous post. Starting a
> large motor does require a large alternator (or copper) to supply the VARS
> (current) since the motor PF is so low. Once the motor is turning at
> speed, it's PF is around 0.8 so then power is dominant. In other words,
> current and voltage are severely out of phase with each other during
> starting. The power vector is small and gets larger as the motor spins up.

Not quite. True, the current and voltage are severely out of phase (low
pf), but because the current is so much larger, even with the low pf, the
total *power* drawn is still much higher than normal running. The very
large current at low pf makes for large power, and the large power is what
accelerates the motor rotor so quickly.

> Well, this is a broad statement. It also depends on the type of motor and
> winding. Also keep in mind that with high currents, I^2 x R losses are a
> component that eats power in both the generator winding and other wiring.
>
> Maybe I'm wrong... I may be twisting my head around this to much from an
> electrical rather than mechanical angle. Anyone else care to comment?
>

I think you're pretty close to having it. Just think that if the pf is 1/4
the normal running value (your 0.2 versus 0.8), but the current is 5 to 7
times larger, then the total power is higher than when running (0.2*6I*V >
0.8*I*V).

You're right that a lot depends on the type of motor/winding as well as the
type of mechanical load connected to the motor. But what we've both said
above generally holds true for most induction motors. There are some motors
that will have different starting/running torque characteristic, and lower
starting current surges, but they are usually some special duty types
(wound-rotor, soft-start, etc...).

And what constitutes 'unloaded' for starting a motor isn't always obvious.
A piston compressor moves a certain volume per revolution no matter what.
But if the discharge is wide open at low pressure, it starts easier (hence
why A/C and refridgerators don't like to restart until the discharge
pressure bleeds off). But a centrifugal pump starts easiest with the
discharge shut off (less flow through the pump means less power needed by
the pump shaft). For conveyors and hoists, the load is directly
proportional to the amount of material on the belt/hoist when starting.

>
> This is an interesting example. There is something more going on here.
> Like you are using the engine and alternator limitations (impedance) in
> lieu of the usual motor starting methods to control starting current. Do
> you start this large motor using the public utility grid? It would have
> to have a sophisticated control system to prevent overcurrent and unsafe
> conditions since the public utility is considered an "infinite bus"
> (ideally capable of supplying whatever you ask of it).

We normally prefer to start it from the DG. This is because we have a
special sort of 'demand peak' billing structure with the utility. The *max*
kw recorded during a billing cycle (even just for the few seconds of
starting this monster) causes an extra surcharge. We coordinate the motor
starting whenever possible with the required 'run-time' testing of the DG,
then once it's started we transfer back over to grid power and shutdown the
DG.

Surprisingly, this particular motor (4160V, 3phase, 60hz, 3000 hp) is
started 'straight across the line' by just shutting it's supply breaker. It
has a time-delay in the overcurrent/overload circuits that blocks any
overcurrent trip for the first 30 seconds after starting, to prevent false
tripping. Similarly, the under-voltage and under-frequency protection trips
on the machine are also blocked for the first 30 seconds. Just to prevent
false-tripping when starting on the DG.

The first few times we started it on the grid, the remote substation's
'distance relaying' saw the motor as a fault and tripped the whole
substation. The utility had to fine-tune the distance relaying to not 'see'
past our switchyard into our switchgear and the motor.

daestrom

philkryder

2005-12-17, 6:21 pm

Is it possible someone could walk me through the calculations for a
fictious 10kw generator with the following:
10 KW continuous at 1800 rpm
88 pound -40 kg rotating mass at
4 inch 10 CM average radius.
5% max cycle drop under surge load.
how much power is available by drawing down the rotating energy during
a 10 second motor start load?

thanks
Phil

Pop

2005-12-18, 11:21 am

Wow, I've worked with a lot of different things, but never
anything on that scale! I'm jealous <g>.

Pop

meow2222@care2.com

2005-12-18, 11:21 am

philkryder wrote:
> I've seen several posts that relate the ability of a generator to
> handle a surge load to the stored rotational energy in the engine and
> generator itself.
>
> This has caused me to wonder about a few things.
>
> How important is the stored rotational engergy compared to:
> 1) governor responsiveness
> 2) Torque rise of the engine.
> 3) Generator design
>
> Specifically, how much energy IS stored in a typical generator running
> at 1800 rpm versus how much is stored in a lighter generator running at
> 3600 rpm?
>
> Does the energy stored go up with the square of the rotational speed?
> Does the enegy stored go up linearly with the mass of the rotating
> components?
>
> Also, how important is constant voltage - versus constant cycles when
> trying to start a reactive load?

I've used small gens, 3 - 7.5kW, and the v and f drops during load
starting are large, typically to well below half the usual genny speed.
They tend to dog right down. So youre miles out of spec on both v and
f.

The field excitation drop is a real advantage with these small
machines, as it drops the V_out, thus reducing the electrical load, and
enabling the genny to keep running while the motor soft starts. If it
weren't for this, the engine would stall rapidly.

ISTR someone mentioning 5% regulation: on these sizes of machine, that
would be very optimistic, for V or f even during run time.

NT

philkryder

2005-12-18, 2:21 pm

me too!

It sounds like a fun job!
Perhaps a 3616 cat...

Anyone up for walking through the hypothetical scenario above?

daestrom

2005-12-18, 4:21 pm

"Pop" <nobody@devnull.spamcop.net> wrote in message
news:JPKdnbFx3cPz7DjenZ2dnUVZ_t6dnZ2d@usadatanet.net...
> Wow, I've worked with a lot of different things, but never
> anything on that scale! I'm jealous <g>.
>

Since I'm also a railroad fan, it helps when reading through the EMD
manuals. Lots of paragraphs that start out with, 'For road use ...., for
stationary use....'. The DG's are actually railroad locomotive engines
mounted on a stationary bedplate with a 60hz generator substituted for the
normal railroad generator.

daestrom

SolarFlare

2005-12-18, 4:21 pm

Not quite. Daestrom would be correct on this one. Power
(or wattage) is used when real work is performed.
Accelerating a motor from stalled to full speed takes
real power or wattage and the PF during that time would
be close to unity, very good or 100%.

When the motor is up to speed, if it is just spinning
and doing no work, now your PF goes down to almost 0
(all VARS) as there is no actual work being performed.
What you have is a bunch of inductive coils with AC
voltage on them doing next to nothing.

Don't be reluctant. It can impede your progress and
choke your growth. (groan)

Motor principles state:

"One good turn, deserves another"

"RF Dude" <post@thisnewsgroup.com> wrote in message
news:uIIof.5518$PQ3.846981@news20.bellglobal.com...
> See below:
>
> "daestrom" <daestrom@NO_SPAM_HEREtwcny.rr.com> wrote
in message
> news:4ZGof.31598$XJ5.20014@twister.nyroc.rr.com...
'large reactive' load.[color=darkred]
cycle or two (for[color=darkred]
But a motor can draw[color=darkred]
long as it takes to[color=darkred]
load, that can be as[color=darkred]
heavy load, it could be[color=darkred]
>
> Your comment conflicts with what I stated in my
previous post. Starting a
> large motor does require a large alternator (or
copper) to supply the VARS
> (current) since the motor PF is so low. Once the
motor is turning at speed,
> it's PF is around 0.8 so then power is dominant. In
other words, current
> and voltage are severely out of phase with each other
during starting. The
> power vector is small and gets larger as the motor
spins up. Well, this is
> a broad statement. It also depends on the type of
motor and winding. Also
> keep in mind that with high currents, I^2 x R losses
are a component that
> eats power in both the generator winding and other
wiring.
>
> RF Dude
>
>

Tony Wesley

2005-12-18, 5:21 pm

SolarFlare wrote:
> Don't be reluctant. It can impede your progress and
> choke your growth. (groan)

You seem wound up. You need some relaxation.

daestrom

2005-12-18, 5:21 pm

"philkryder" <alt.google@Kryder.com> wrote in message
news:1134854911.452336.82070@f14g2000cwb.googlegroups.com...
> Is it possible someone could walk me through the calculations for a
> fictious 10kw generator with the following:
> 10 KW continuous at 1800 rpm
> 88 pound -40 kg rotating mass at
> 4 inch 10 CM average radius.
> 5% max cycle drop under surge load.
> how much power is available by drawing down the rotating energy during
> a 10 second motor start load?
>

Well....

Your 10 cm average radius isn't quite the right thing to use for calculating
moment of inertia. Ideally we could weigh each part and use its geometry to
find the 'I' value for each and sum them all together for the total.

But, just to take a 'shot' at it. Suppose the moment of inertia might be
similar to that of a solid cylinder that has a radius of your 10 cm. So the
moment of inertia ('I') would be...
I=1/2 * m * R^2 = 2000 kg-m^2

Rotating at 1800 RPM (30 RPS), its kinetic energy would be...
K.E. = 1/2 * I *omega^2 = 1/2 * 2000 (kg-m^2) * (30 /s)^2 = 900000 Joules

After it slows to by 5% to 28.5 RPS, its kinetic energy would still be...
K.E. = 1/2 * 2000 (kg-m^2) * (28.5 /s)^2 = 812250 Joules

Meaning we could extract the difference of 87750 Joules. That's equal to
87750 watt-seconds, so if we withdrew that energy over 10 seconds, we could
draw out 8775 watts for those ten seconds.

Mind you, the actual moment of inertia is probably less than 2000 kg-m^2.
If you have more details of the rotating elements, we could try to get a bit
closer.

Frankly, these numbers seem high to me. But that's the math. If the
generator rotor was 20 kg with a 10 cm radius (I=1/2*20*10^2 = 1000 kg-m^2),
and the crankshaft was the other 20 kg with a 6 cm radius (I=1/2*20*6^2 =
360 kg-m^2), the total moment of inertia would drop by a factor of 1360/2000
and the answer also would drop to 5967 watts. So you can see getting the
'I' right is the 'devil in the details'.

If we change just the synchronous speed of the system to 3600 RPM (60 RPS)
and drop it's speed by the same 5% to 57 RPS, we have an initial KE of 3.6e6
J, dropping to 3.249e6 J for a delta of 3.51e5 J. Over 10 seconds that
would represent a power of 35.1kW. See how doubling the speed means a lot
more stored energy?

That's not to say that 1800 RPM engines are a bad choice. Lower speed
usually means longer life. And diesels usually have heavier parts than gas.
So a hi-speed gas engine, with lightweight parts that are just barely strong
enough to do their job may not handle surges as well as a lower speed
diesel, with heavy duty parts that have a significant strength margin. I've
seen a few high speed single cylinder gas engines that have thrown a rod or
cracked a crank when suddenly loaded with a large load. The only diesel
I've ever seen with a broken rod was caused by trying to start it with oil
in the cylinder (leaked in from a pre-lube system and a faulty valve seal).

Now, even if the governor didn't react at all during your 10 second motor
start, the engine output power may increase anyway, owing to the slower RPM.
And the existing load (if any) on the generator may drop slightly with the
speed reduction, leaving a slight power imbalance to favor handling a
slightly larger surge.

And with piston engines, there is a problem in that you can't use the weight
of the rod/piston mass (well, at least not all of the rod). These represent
'reciprocating weights' and although they certainly move, their kinetic
energy is constantly changing because the piston slows and stops at each end
of travel. One end of the rod moves exactly with the piston, while the
other end rotates with the crank. One rule of thumb is to use 1/3 the rod
mass at the radius of the crank pin for its 'I'. If the engine has multiple
cylinders in a 'V' bank (or multiple cylinders but the crank pins offset
from 180), you can take *some* credit for the reciprocating weights since
they don't all stop at the ends of their stroke at the same time, but that's
beyond me.

daestrom

meow2222@care2.com

2005-12-19, 1:21 am

daestrom wrote:

> If we change just the synchronous speed of the system to 3600 RPM (60 RPS)
> and drop it's speed by the same 5% to 57 RPS, we have an initial KE of 3.6e6
> J, dropping to 3.249e6 J for a delta of 3.51e5 J. Over 10 seconds that
> would represent a power of 35.1kW. See how doubling the speed means a lot
> more stored energy?

IRL a faster engine needs much less flyweel mass, so has less mass.

NT

philkryder

2005-12-19, 1:21 am

Wow!

Thanks for all the effort.

I wish that I had some "real" numbers to contrast a 10kw 1800 RPM
gaseous unit with a 3600 RPM air cooled.

It is interesting that it takes only 1/4 the moment of inertia for the
3600rpm to match the 1800 rpm unit.

I too am guessing that these numbers are HIGH - otherwise the
advertizers would be touting massive surge capabilities.
And, there would be someone out there adding circumference weighted
flywheels to exagerate the effect...

So, the two critical items -
the mass
and the distance from the crank centerline are both unknown in the
real world.
Any real world data on 10kw class units would be appreciated.

Thanks for your efforts
very interesting.
Phil

LinkBot