Author Topic: Peak output and flywheels  (Read 25244 times)

Jim Mc

  • Sr. Member
  • ****
  • Posts: 330
    • View Profile
Re: Peak output and flywheels
« Reply #30 on: August 16, 2006, 02:51:59 AM »
Right, akaik, they only showed up on early the DJA sets (single cylinder Diesel, 1800 rpm.)

Interesting you mention the CCK sets.  I have an old one here, and the engine has a lot of blow-by.  Probably not worth fixing.  But I'd been thinking that its 1800 rpm alternator might make a nice mate for a  listeroid.  Biggest problem would be adding a 2nd bearing.  The nice thing is that on this set, the alternator has a 12V DC starting motor built in. 

I have a schematic showing the flicker point scheme, but it's a pdf.  What's the best way to post it ?


Jim Mc

  • Sr. Member
  • ****
  • Posts: 330
    • View Profile
Re: Peak output and flywheels
« Reply #31 on: August 16, 2006, 02:56:16 AM »
The capacitance cancels some inductance in the field winding, it does not make much difference how long it stays in the circuit it's job here is not to deliver a pulse of energy but to continiously cancel winding inductance and increase current flow.

Kindly show us how that works.  I can't fathom it.  There's a heck of a lot of things I don't understand in this world.  But with 23 years' experience designing electronics for a living I ought be able to get this!




mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Peak output and flywheels
« Reply #32 on: August 16, 2006, 03:03:04 AM »
Jim: the reason the onan at 3600 does not have the issue of flicker is one of resolution

the lower the rpm of the engine and the larger the step up in gearing the resolution works against you

a small change in engine rpm relates to a relatively larger change in frequency compared to a direct drive 3600 rpm genset

if you drop 30 rpm on a 650 engine, that might equate to a 5 hz drop,

the same 30 rpm drop on a 3600 rpm engine is likely less than a hz.

i dunno without doing the math,,, but i have experienced this phenomenon on
home built light plants

the best results i ever got was running the engine at 3600 and gearing down to 1800 or in some cases 1200, thus increasing the resolution and giving a much more stable frequency.

make sense?

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

Doug

  • Guest
Re: Peak output and flywheels
« Reply #33 on: August 16, 2006, 03:03:09 AM »
I dunno but I want to see that print, I don't remeber this flicker crap at all.

The CCK head was if I remember something on the order of 4 or 5 kw. These heads also could motorise and start the engine. Pitty about the engine.
I'll say this again for like the umpteenth time. Indian heads could be adapted to start an engine of somebody over thre would just add a com.....

Doug

Jim I think he;s talking about something that would look like series resonance for a split second...

Jim Mc

  • Sr. Member
  • ****
  • Posts: 330
    • View Profile
Re: Peak output and flywheels
« Reply #34 on: August 16, 2006, 03:37:39 AM »
...something that would look like series resonance for a split second...

Still doesn't make sense to me.  Series resonance involving what inductance - the field???

The pdf I have is my own design, not Onan's.  I don't have a manual that covers the DJA sets, but I'd like to see it myself.

Yes, Bob that makes sense.  The flicker problem gets worse with slower speed engines, single cyliner engines, and with Diesels since the energy required to get past TDC on the compression stroke is greater than with a gas engine.



Rtqii

  • Full Member
  • ***
  • Posts: 152
    • View Profile
Re: Peak output and flywheels
« Reply #35 on: August 16, 2006, 03:42:40 AM »
The capacitance cancels some inductance in the field winding, it does not make much difference how long it stays in the circuit it's job here is not to deliver a pulse of energy but to continiously cancel winding inductance and increase current flow.

Kindly show us how that works.  I can't fathom it.  There's a heck of a lot of things I don't understand in this world.  But with 23 years' experience designing electronics for a living I ought be able to get this!

No problemo  8)

Let's start with a review of capacitance and inductance.  These two values cancel each other out. If the value of the capacitor exactly cancels the inductance of a coil... As far as the resulting circuit is concerned, only resistance remains. This is the basic principal of the LC oscillator.

http://www.electronixandmore.com/articles/oscillators.html

http://en.wikipedia.org/wiki/LC_circuit <- Da math

Now, if you put a capacitor value across the rotor that canceled the inductance completely, you would end up with a resonate circuit that would increase the harmonic distortion of the generated energy. This would be the result of using the large capacitor you described on an ST head, and on an iron cored coil such as an ST rotor you are probably correct in that it would take a huge capacitor value to cancel the inductance completly. Yet... This is not what we want.

The lightbulb in my head went off when you mentioned that with a cam driven set of points, the capacitor would only be in the circuit for a fraction of the engine rotation period, and it would be closed in the circuit only when a voltage boost was required to overcome flicker.  During the period of time when the capacitor would be in the circuit a complete canceling of the inductance is not required, we would only need to cancel enough inductance to boost the current flow thru the rotor and preventing voltage from dropping.

With an inductance meter measurement of the rotor inductance, and a scope measurement of the voltage drop causing the flicker, you can calculate the value of the capacitor required to boost the rotor current by a given percentage... Let's say a 10% voltage drop is resulting in flicker, you cancel 10% of the rotor inductance (a reasonble sized capacitor) and switch that in with cam operated points...

Now, with that in the can... Let me qualify this by saying I have never done something exactly like this... I have done my fair share of PFC math in order to completely cancel inductance in power processing applications (the BIG capacitors you described)... And the exact same principal is applied to the primary coil of the old style point driven spark ignition systems... It is also the basis for LC oscillators... All of these applications I am very familiar with. In this particular application you don't want or need enough capacitance in the circuit to create a completely canceled condition of the inductance (aka PFC correction in a power circuit or LC oscillator functions) you just want enough capacitance switched in to correct the voltage drop resulting in flicker.

Jim Mc

  • Sr. Member
  • ****
  • Posts: 330
    • View Profile
Re: Peak output and flywheels
« Reply #36 on: August 16, 2006, 06:55:31 PM »
Still doesn’t add up in my head, Rtqii.  You describe a parallel LC circuit, like PFC caps, right?  Whenever I’ve used PFC caps, the load current (the current through the inductive element) doesn’t change for the addition of the caps.  Of course the line current (the sum of the load and capacitor current) drops, but the inductor current remains the same.

PFC caps aside, here’s the way I look at it, and the reason I think caps won’t work:  The fundamental 60 Hz AC output from the stator (which is what we want to boost, right?) is a function of the DC flowing through the field. The caps are an open circuit at DC, and therefore have no effect on the DC component.  Therefore, I can not see a mechanism by which adding capacitance across the field will change the DC component of field current.  If you can’t change the DC component of the field current, you won’t boost the 60 Hz AC output.

Doug

  • Guest
Re: Peak output and flywheels
« Reply #37 on: August 16, 2006, 07:53:07 PM »
Jim I don't know now....

I need to see a print, and my manuals for the Onans are at camp ( where the Onans are... ). I thought I gasped what he was trying to explain and it seemed to make sence to me but I was also enjoying some cool refreshing beer.

Doug





Rtqii

  • Full Member
  • ***
  • Posts: 152
    • View Profile
Re: Peak output and flywheels
« Reply #38 on: August 16, 2006, 08:44:09 PM »
Still doesn’t add up in my head, Rtqii.  You describe a parallel LC circuit, like PFC caps, right?  Whenever I’ve used PFC caps, the load current (the current through the inductive element) doesn’t change for the addition of the caps.  Of course the line current (the sum of the load and capacitor current) drops, but the inductor current remains the same.

PFC caps aside, here’s the way I look at it, and the reason I think caps won’t work:  The fundamental 60 Hz AC output from the stator (which is what we want to boost, right?) is a function of the DC flowing through the field. The caps are an open circuit at DC, and therefore have no effect on the DC component.  Therefore, I can not see a mechanism by which adding capacitance across the field will change the DC component of field current.  If you can’t change the DC component of the field current, you won’t boost the 60 Hz AC output.


Well... Let's table this discussion until I get my engine and generator and can properly document an experiment. My meters and scope are in storage waiting until the new building goes up, and I just ponied up for the engine and generator today. This is something that won't take a great deal of effort to prove or disprove, once I get my infrastructure in place.

The thing I am looking at in the circuit function I see, is that the DC is not steady state... Not like a battery output. The rotor coil is not acting like a pure permanent magnet... There is magnetic flux in the rotor because voltage is rising and falling with the frequency lags during portions of the engine cycle. I honestly think some capacitance switched in during these lags will result in smoothing the output.

The proof of course is in the pudding... It will take me a few months before I can come back to this with the proper equipment and run some tests, but I don't mind looking into it further then.

rcavictim

  • Certified Generator Head and Grand Master Sparky
  • Hero Member
  • *****
  • Posts: 1827
    • View Profile
Re: Peak output and flywheels
« Reply #39 on: September 19, 2006, 06:53:51 AM »



^^^ Frequency stability and load starting. This is my working design for an 1800 RPM jackshaft with steel flywheels. The shaft would be belt driven from the engine, and the generator would be direct coupled to one shaft end. The flywheel mass and higher rate of rotation should really improve the "flicker" quality of the input signal and provide torque on demand at a high rate of delivery to energize my physics loads.



Rtqii,

The high speed flywheel pack directly coupled to the shaft of the alternator is a very good idea.  In addition I would highly recommend oversizing the kVA rating of your alternator by a factor of 2X, 3X or more, i.e. use a minimum 40 kW unit if the diesel engine can only produce 20 kW continuous electrical output.  The lower Z will really stiffen your supply.

In a multi kVA and larger Tesla coil oscillator with rotary break the use of a break synchronous with the supply frequency really pounds on the alternator and requires an alternator well oversixed for the load.  You can get by with a more conservative power source if you stay with higher PPS bang rates (than the mains frequency) like 400-600 PPS and non synchronous operation.  High rep rates are much harder on pulse capacitors however and shorten their lifetimes.  Engineering is full of tradeoffs.  Of course it all depends on what you are trying to achieve.
-DIY 1.5L NA VW diesel genset - 9 kW 3-phase. Co-gen, dual  fuel
- 1966, Petter PJ-1, 5 kW air cooled diesel standby lighting plant
-DIY JD175A, minimum fuel research genset.
-Changfa 1115
-6 HP Launtop air cooled diesel
-Want Lister 6/1
-Large DIY VAWT nearing completion

Rtqii

  • Full Member
  • ***
  • Posts: 152
    • View Profile
Re: Peak output and flywheels
« Reply #40 on: September 19, 2006, 06:42:19 PM »
Of course it all depends on what you are trying to achieve.

Maximum peak power of course!!!  ;D

I tend, based on experience, to think you are very much correct about oversizing the generator set. I know for a fact how hard these circuits are on everything.  No sooner do you beef up the weakest component in the system (because the previously engineered component failed in serivce), then the next weakest component in the system fails... When the step up transformer, tank circuit and gaps are all pretty much bullet proofed, the next place to look for a failure would logically reside in the control cabinet & power supply.

I am hoping to keep the actual design, research, and prototyping power levels in the range between 3-6 kW, and figured on installing an ST15kW generator head for the mains supply. I could do an ST20, I held back at the last second on the genhead purchase because I have no place to store it until the shop/lab building is roofed and that is a few weeks away yet. There is a lead time in importing the engine, so I paid for that and held the check for the ST15 back.

There is the inevitable temptation in this line of research to match a larger engine to a larger generator and anybody familiar with this equipment knows there is an infinity quotient directly associated with Tesla's resonance patent... There is no limit to how large and powerful this equipment can be scaled up to. It can lead to bankruptcy.

I have designed, built, and operated Tesla oscillators and resonators in the 10-15 kW range, and having done this I am of the opinion that this is 2-3x above the power levels needed in order to produce viable prototypes and demonstrations of the new applications I have patented.  Most of the time I think 3 kW of input power should do the trick, and if it doesn't then 6 kW surely will make my points.

I have liquidated, or am in the process of liquidating, practically all my assets to put this project together... I have the money required to do the job right, but the sky is not the limit. With this in mind, I sized the power plant for dual use... It will run the ranch most of the time, and it will power the research for a small fraction of the time. Giving the balancing act required whenever a project like this is designed, engineered, and paid for... I opted to go with a 20 HP prime mover to turn the generator, and a 15kW ST. I would not consider spending a few hundred extra on an ST20 to be extravagant.

The goal here with the Tesla research is to design, engineer, and prototype patent models for commerical development... Not the commerical applications themselves. Admittedly, I could probably do this work on a much smaller scale, and could even use solid state oscillators... But it would not give people the picture of what a commerical type installation and application would look like... In other words, the proofs can be done on apparatus you can probably hold in your hand and plug into a wall outlet... But in my experience people have limited imagination and are not able to see that scaled up equipment operates on exactly the same fundamental physics, only everything is larger.

So in effect what I am doing is developing a research facility as a scaled down mini-plant... Fuel will come in, drive the prime mover, spin the jackshaft with flywheels and turn the generator.... It will have all of the basic components of a stand alone commerical enterprise, just on an R&D scale.

Good capacitance is expensive... I have some good capacitance with low minutes, and it has never been excited at anywhere near its rated voltage and capacity... They were designed to be tough enough to process 20 kW all day, and should have 72 hours of life in them at 40,000 volts AC rapidly pulsed. I should be able to get many, many research hours out of them at much lower voltages and power levels.

If I were to decide to go with a much larger generator, I would end up moving to higher power levels. If I were to do this I would convert the main oscillators over to 3 phase and pick up a used commerical 3-phase generator to drive it. I suppose this is possible eventually, I have thought about it, but it is not in the cards at this point.  I think perhaps what I may do is put some extra floor space in the engine room, leave a blank spot in the floor where a mounting block for this upgrade could be placed in the future... Then, if I see a steal deal on a big 3 phase generator...

But I would not remove the Listeroid and ST genset from service... I still need a primary power supply for the ranch.

« Last Edit: September 19, 2006, 07:06:24 PM by Rtqii »