Author Topic: Changfa Slow Speed Experiment  (Read 43771 times)

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Changfa Slow Speed Experiment
« Reply #45 on: April 29, 2009, 03:49:00 AM »
thanks Jim:

always good to have input from the guru"s

another question if i might

i get the resistance of the st head, ~.1 ohm, another .1 for the wire run, but
are we not neglecting the resistance of the load?

ih any event the losses as you surmise at around 140 watts of 5% doesn't seem like much to be worried
about unless one is running long hours? and is worried about eeking out efficiencies?

the reason i ask is based on the cumulative effects, if we discount a percent here, a couple there, another over
there,, pretty soon we have some losses that begin to cost fuel to cover?

anyway thanks a lot for the explanation, and forgive me for being a bit thick between the ears
i am under the weather pretty bad for the last week with the flu,, thankfully the garden variety type
and not the pig kind!

if you don't mind i might (probably) will have other followup questions, later when i feel better.

thanks

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

Jim Mc

  • Sr. Member
  • ****
  • Posts: 330
    • View Profile
Re: Changfa Slow Speed Experiment
« Reply #46 on: April 29, 2009, 03:59:49 AM »


another question if i might

i get the resistance of the st head, ~.1 ohm, another .1 for the wire run, but
are we not neglecting the resistance of the load?

Right, it's neglected.  The load resistance does not enter into the calculation of the power loss due to a low PF. The load gets 2.6kW in either case.



mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Changfa Slow Speed Experiment
« Reply #47 on: April 29, 2009, 06:36:41 AM »
not wanting to argue a point i am unsure of, i need to dig back into the text again

iirc from a few days ago my reference's state that the resistance of the alternator, wiring (along with all interconnects, breakers,
twiston's etc) and the resistance of the load (reference was to motor loads and maybe different for others?) all must be include in
figuring losses.  i gotta confirm that for sure.

i don't know if it is worth fighting over or not, striving to capture but
i do know i can measure the difference in fuel consumption a 50watt difference in  load fairly easily in fuel consumption.

i think the only way i will ever have a difinitive answer is to do some actual fuel consumption tests with a fixed load at X pf
, then correct to unity and retest again to see what the results might actually amount to, and whether of not the difference's are
significant enough to go after?

going back to the OP i would toss this out
at 2.6 kwatt load (whatever the powerfactor) and at low speed
the fuel consumption in grams/kwatt/hr will be higher than if
the engine was allowed to run at 1800 rpm and produce~7kwatt

in all my tests the 195 changfa driving an st head just can't compete favorably at this low a loading
with a 6/1 according to my results.

now a 195 driving a specially prepared 110-555 into a 24volt nom battery bank can compete very well with a 6/1
at ~2.9kwatt output (being dc pf is not an issue) at lower engine rpm.

so i guess it really depends on what the end use or goal is for this 195 low speed project?

(i gotta go back and reread to find out what the original plan was)

:)

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Changfa Slow Speed Experiment
« Reply #48 on: April 29, 2009, 06:49:22 AM »
i almost forgot one of the most important aspects of trying to get max output from the changfa
and this will be aggravated at lower speeds.

if the engine is a idi changfa it will have 20:1 comp ratio
at max loading of around 7.3 kwatt at 1800rpm you will start to have head gskt fire ring failures
with the oem gasket, the fire ring is too thin and too hard at the fold line. it will crack and fail, leaking
compression into the coolant which will show up upon the next cold startup.

if you calculate 2.6 to 2.8kwatt as max loading for ~1000rpm you may be in this stress region for this gskt

one has to remember the injection timeing is optimized for 2000rpm at full rated load
half that speed and the timing might be a couple degree's too fast, this adds siginificantly to the cylinder pressures
and likely will lead to gskt failures.

all is not lost though, there is a better gskt available :)

just wanted you to be aware of the risks

i have cracked several oem changfa fire rings, doing hard run's (equivalent to dyno torture tests) on my changfa

of note, the rest of the engine could care the less how hard i beat on it
it appears the head gskt is the weak link, and maybe a nice sacraficial one at that?

cheaper to kill a gskt than beat a rod to death i suppose?
(although i think it would take a lot more to kill any hard part in a changfa)

just wanted to pass that along

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

Veggiefuel

  • Energy Machines Ltd.
  • Full Member
  • ***
  • Posts: 163
    • View Profile
    • Energy Machines Ltd.
Re: Changfa Slow Speed Experiment
« Reply #49 on: April 29, 2009, 03:45:06 PM »
Bob,

Thanks for bringing up those important points.
If I am going to design a system built around a slow running Changfa, then those points must be addressed.

My engine is a direct injection version. The manual states a 20:1 ratio.
Would  you still expect the same problems with a DI engine ?

1] You noted that a better head gasket is available. Can you advise where this can be purchased?

2] Injector timing:  I looked in my Changfa manual on this subject but the translation from Chinese to English became enough of an issue to make it dangerous. ;)
Have you played around with your timing?, if so, what's the best method of adjustment.

3] How do you know when a fire ring cracks??

Thanks for your comments,
Veggie
GM90 engines, Changfa's, Voltmaster Generators, Pellet Mills - www.energymachines.ca

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Changfa Slow Speed Experiment
« Reply #50 on: April 29, 2009, 04:05:44 PM »
iirc the DI changfa engines are ~18:1 or lower comp ratio (the exact i don't remember but lower than the idi engines)

a better gskt is available from gasket to go, ask for the changfa 1100 gskts "if" you engine is truely an idi and not  a di engine.

the DI engines use a different head gskt that looks to be heavier built to start with

injection timing is altered by changing gskt shims under the injection pump, iirc it take .004" to alter 2 degree's

i have not tried to alter the timeing for running at slow rpm, my primary use is at 1800rpm at heavy loading and
at 1200 rpm at low loading, so i have not looked into changing the timing.

you won't know when the fire ring cracks unless you are very well instrumented, then you might notice a slight change in power
while running, however
the next time you try a cold start, you will have water forced out the overflow of your cooling system, hard starting if at all
and exhaust vapor clouding.

bob g



otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

trigzy

  • Sr. Member
  • ****
  • Posts: 313
    • View Profile
    • High Tech Hicks
Re: Changfa Slow Speed Experiment
« Reply #51 on: April 29, 2009, 07:53:57 PM »
not wanting to argue a point i am unsure of, i need to dig back into the text again

iirc from a few days ago my reference's state that the resistance of the alternator, wiring (along with all interconnects, breakers,
twiston's etc) and the resistance of the load (reference was to motor loads and maybe different for others?) all must be include in
figuring losses.  i gotta confirm that for sure.
Those can all cause losses.  Some more than others, ie. undersized cables, bad connections.
Quote
i don't know if it is worth fighting over or not, striving to capture but
i do know i can measure the difference in fuel consumption a 50watt difference in  load fairly easily in fuel consumption.
If you can measure a 50 watt load by fuel consumption, I'm impressed.  You would be easily be able to remove 50 watts of load, by correcting the bad PF at the source(s).  That is the only way to remove the losses that the excess current is causing in the cabling system.  Otherwise you are just removing those loses from the windings, and might not see that much of a difference.
Quote
i think the only way i will ever have a difinitive answer is to do some actual fuel consumption tests with a fixed load at X pf
, then correct to unity and retest again to see what the results might actually amount to, and whether of not the difference's are
significant enough to go after?
 
Do you have equipment to accurately measure PF on bigger loads??  That is to say, something with a higher power rating than a Kill-A-Watt?  If so, I'd be interested in hearing about it.  Please don't tell me you are going to correct the pf based on the motor nameplate alone.  Note that a motor will have a different pf loaded vs. unloaded.  So don't measure the pf of your circular saw spinning freely, and then expect the same system to correct the pf if loaded up.  On a 2.7kW load, I'd bet you'd easily save the equivalent of 50 watts of fuel by correcting the power factor, unless its already better than 0.95.  However, when you correct the pf, you might find the voltage will be higher, so resistive loads might draw more power.  So you'd be saving fuel on waste, but the other loads might consume more.  As long as they were useful loads, you would be getting it back in appliance performance (example: water heater elements would draw more power, but wouldn't run as long).

Sorry if I was re-iterating something you already knew, but thought I'd also post it for the benefit of the group.

Steve
Power Anand 24/2, Brushless 20kW, some other antique iron.
Vendor of AVR's, Small Clones of Yanmar Diesel and Honda Gasoline Engines

trigzy

  • Sr. Member
  • ****
  • Posts: 313
    • View Profile
    • High Tech Hicks
Re: Changfa Slow Speed Experiment
« Reply #52 on: April 29, 2009, 09:22:43 PM »
You would be easily be able to remove 50 watts of load, by correcting the bad PF at the source(s).  That is the only way to remove the losses that the excess current is causing in the cabling system.  Otherwise you are just removing those loses from the windings, and might not see that much of a difference.

Ummmm ..... I am not sure if I understood you right but just for the record, the power factor correction should happen at the load and not the source IMHO. The power source will then see a pf of 1.0 and you won't have losses from the source to the load. I would think that the load itself would still have excessive currents circulating. IE the power factor correction will not do anything to help with pf inefficiencies in the motor de to the power factor .

.... at least based on my interpretation and understanding.

Jens
Jens,
      Sorry, it made sense in my mind...   I meant at the source of the bad pf, ie. at the load with the bad power factor.  Not the source as in generator.  My apologies for clumsy wording.  The inefficiencies would then be limited to that load, as you stated above.  Also, I forgot to mention that you'd want to switch the pf correction with the load, so that you don't end up with leading power factor, instead of the lagging power factor you were trying to correct.

Steve


Power Anand 24/2, Brushless 20kW, some other antique iron.
Vendor of AVR's, Small Clones of Yanmar Diesel and Honda Gasoline Engines

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Changfa Slow Speed Experiment
« Reply #53 on: April 30, 2009, 01:03:47 AM »
i can measure pf all the way up to around 200amps, if needed
but of course i can't do anything near that with the st7.5

in other words i can get accurate numbers on any st head i am likely to need to test a load against.

and yes i am aware that pf shifts depending on loading
an unloaded induction motor can have horrible pf, one 1/2hp motor i have registers .26 pf no load
but only improves to about .4 under moderate load,, and haven't checked under heavier load yet.

my 8 inch bench grinder is similar, it starts out at .36 and improves somewhat when grinding
but not nearly what i would expect doing normal grinding.

thanks guys

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

trigzy

  • Sr. Member
  • ****
  • Posts: 313
    • View Profile
    • High Tech Hicks
Re: Changfa Slow Speed Experiment
« Reply #54 on: April 30, 2009, 01:58:07 AM »
Ok,
    Let us know what you find, and any improvements you make.

Steve
Power Anand 24/2, Brushless 20kW, some other antique iron.
Vendor of AVR's, Small Clones of Yanmar Diesel and Honda Gasoline Engines