My "perky\cat/volvo" DC bat charger

[mobile_bob]

Lloyd:

i finally got it to load after about an hour, i gotta get ahold of comcast and get my cable connection sorted out!

as for my way of "injuneerun", well it is really no different than that of the early developers or engineers of any product
wherein they run all sort of test, then from those test results compile formula that explains what there results are,

they then use those formula to design and build the next generation, then redo all the testing again to see if the results of the
new machine match what the formula predicted, if so then they have a useful formula to go forward and work with, on the other hand
sometimes things don't come out as predicted by the formula and thy are left to sort out why.  was the formula flawed? or was there
some other factor in play that they had not accounted for? or is there something other that they overlooked? etc.

this is how the formulas that we use daily came to be, from empirical testing, save for those formulas that are theoretical in nature, those
are called oddly enough "theoretical laws" until such time a method of testing can be determined that will either prove or disprove the theory.
if proven then the theoretical law, becomes a law or formula.  at least that is how i understand it.

just yesterday i watched an interesting show on the "science" channel on the development of the orion lander projects heat shields
they had the engineers of all stripes modeling on their puters all facets of the project, incredible amounts of puter power and guys with
more smarts in one finger than i have in my whole body working on the problem,, using formulas laid out from empirical data, testing and results
from the appollo project 40 plus years ago.  Then it was up to the test engineers to arrive at a test to determine if all the engineering was correct.

they ran the test using superheated nitrogen at 10kpsi firing past a 7" model in a vacuum chamber, ran the test only to find that they needed to go back and rework those old formulas to fit the results they were getting with the new stuff.

in a related test the ceramic coating started to burn away at 5k degree's when the formulas said it was to be good for twice that.

so the moral to my diatribe?

if i were to want to design from the ground up a turbo alternator that was 2 pole 3600rpm of several hundred kva or higher, i would definitely use
all the data, engineering and formula's from the books i could find. even for a 4 pole 1800rpm i would do so.

but i am working with 12 pole 500hz clawpole alternators (now) and previously 16 pole 600hz alternators, and i don't have any design text to work with for the clawpole alternators, and...
even if i did i would also need some very specific and in most cases proprietary information on materials used by the oem in order to effectively
use the formulas provided for clawpole machines.

i am not saying that working up front with established formulas are bad, because they are not, but

i need clawpole alternator formulas for 400-600 hz machines,
i need to know from the oem what the magnetic properties of the stator steel are,
i need to know flux densities,
some technical drawing would be useful, etc

none of which i have access to, so
failing that i have to reengineer the product via empirical testing, to get as many data points as realistically possible, so that
i can then go back and see if i can derive formula or see if the existing formula for other machines explain the results of all
that testing.

in the end one of two things will become apparent, either existing formulas will be shown to be effective for engineering clawpole alternators
(i feel this is unlikely) ,or
i will come up with an altered formula that better explains the machines operation under the conditions i am running it at, which will allow me to
make a series of calculations as to what changes will end in a result i am after.

this seems like a solid approach, and is one i have used numerous times when faced with an oem that will not release technical info on their products and where i cannot find text outlining the specific design of the product i am interested in.

i once spent about 100 hrs reverse engineering a drive module of a koyo/automation direct programmable logic controller, because the
oem would not release a schematic or theory of operation. it took a while but i got er done, only to be asked if i would sign a "first right of refusal agreement" with the oem if i were to develop a product based on their plc (which i refused to do, because after spending a hundred hours
without their help i figured they now needed me worse than i needed them)

i digress only to illustrate that what i do is nothing new, it has been done since the beginning of time.

also i will leave you with this

in any formula or rather many formulas there are factors which are called "constants", what i am concerned with is the possibility that
the "constants" used in the formulas for the design of the clawpole type alternators might be different than those of more common
60hz machines. i don't know how else to explain why i can get results that are not predicted by the EE i have talked to. i can only assume that
there is something different about the some factor of the formula and that is likely a constant changes value somewhere.

in transformer design there is a constant that is "4" if the power is a pure sinewave, and the constant changes from "4" to "4.4" if the power is square wave.  that factor alone can make a huge difference in the design of a transformer, so basically if one mistakenly assumed that he should use the 4 instead of 4.4, he might get a unit that does not work as designed and he would be left scratching his head and wondering why.

anyway, if you come across any reference material on the design of automotive clawpole design alternators i would dearly love to get my hands on it.

bob g

[Lloyd]

ARMY TM 9-6115-464-34 AIR FORCE TO 35C2-3-445-2 NAVY  NAVFAC  P-8-624-34
Some 400-600hz mixed

http://www.tpub.com/content/dieselgenerators/TM-9-6115-464-34/index.htm

[Lloyd]

Hi Bob,,

Just back from Blanchard, Todd ran the alt, but the max rpm he could run is 5000. Also he can''t adjust the field voltage...only the field current. He also said it didn't think it wise to exceed 8 amps on the field, so I need to contact KEI and find out what the max field current is on this alt.

So these are the only numbers I could get.

rpm               field  12v/5.5 amps current                phase to phase voltage AC

3000                         5.5 amps                                        119.4 rms

4000                         5.5 amps                                        161.0 rms

5000                         5.5 amps                                        194.0 rms          

It looks like  if I want a full matrix, I will need to have KEI rewind a machine that they can run and fill in the matrix to get the full grid.

Lloyd    

[Lloyd]

Just talked with Glen @ KEI, he's going to rewind an alternator and give me the numbers to fill in my excel spreadsheet above.

I also got some specific info on the the windings both rotor and stator.

The rotor is wound 401 turns 17 guage, and 2 ohms cold. Max voltage on the rotor field is 18 volts.

The stator is wound 2 in hand 12 gauge, 1st in hand is 30 turns, 2nd in hand 24 turns for a total of 54 turns per phase, he was at McDonlads with his kids and couldn't remember the ohms of the stator phase. So he will  also give me the stator ohms once the new machine is wound.

Lloyd

[Lloyd]

The Lundell alternator is the most common power generation device used in cars. It is a wound-field three-phase synchronous generator containing an internal three-phase diode rectifier and voltage regulator.

The maximum alternator output is limited by heating of the rotor and stator windings and by magnetic saturation of the machine. As shown in Figure 2, conventional Lundell alternators are limited to about 2kW. The efficiency of these alternators is about 40 to 55%.

2 THEORY
2.1 Alternator Electrical Model
Figure 5 shows a simple alternator model with a SMR. The stator is modelled as a Y-connected three-phase sinusoidal voltage source with each phase including a leakage inductance. The SMR is essentially a boost converter following a rectifier. The SMR acts a DC/DC converter which allows the effective DC link voltage seen by the machine to be reduced from the actual DC link voltage, VDC, to (1-d)VDC, at a duty-cycle d. This extra control flexibility over an uncontrolled rectifier allows more power to be extracted from the alternator over a wide speed range. This allows the SMR to be modelled as a standard rectifier with variable output voltage. With a SMR, the field current can be kept constant at its maximum value as the output voltage regulation is done by duty-cycle control.

The SMR can be modelled by the machine phase equivalent circuit, as seen in Figure 6. This model is based on the following given assumptions [5], which allow the modelling of the rectifier and voltage source as a variable three-phase resistor:
• The rectifier forces the alternator phase currents to be strictly in phase with the phase voltages of the alternator,
• The phase currents continue to be sinusoidal, despite the non-sinusoidal voltage waveforms. where E is the back-EMF, Ls is the phase inductance, Rs is the phase stator resistance; and RL is the effective load resistance. In addition Vo and Io are the alternator output voltage and current, respectively. This model ignores saliency effects, iron losses and magnetic saturation.

http://www.itee.uq.edu.au/~aupec/aupec04/papers/PaperID82.pdf the White Paper


_________________________
--------------

I'm talking about the TEST BED, not sure about the SRM..yet



I predict that lundell efficiency will be most effected by cooling/heat extraction, so far everything I read leads to marginal gains...but one thing I have found as mention in every WP I read is heat.

When speaking with Glen today at KEI, he suggested that to find the losses of the stator and rotor was put each in an oven and start heating them, measure  the ohms in ten degrees data points, he suggested the R would increase liner almost equal to the temp rise.

There are other associated losses eddy, saturation, hys..... , but he also suggested that each of those losses are compounded by heat...

As an marine electrician...I have to de-rate the amplitude of a wire, based on environmental temp, as well as current and again for number of current carrying conductors in a bundle....so this is all starting to come home to roost.

Glen, also  said ...Lloyd I think you have the proverbial "birds nest on the ground" when I elaborated on my seacooler for my project...a steady diet of 48F turbo air temp.

Lloyd

[mobile_bob]

first of all i would like to state that on one hand i am not crazy about the pdf from australia, because it is based on and a repetition
of perrault and caliskan's earlier work at MIT, however it is attributed properly and credit is given to them.

most all of the data, graphs and charts are from the MIT paper.

what i would like to direct your attention to (along with anyone else that might be interested) is figure 16

all of the curves are based on testing of a single 12volt nominal alternator at various output voltages being the only variable
(all else being the same)

as you can see the alternator running at 6krpm producing 14volts will make about 1.75 kwatts, and consumes about 4.5kwatts input power
1.75/4.5= ~39% efficient  which is pretty typical and dreadful in my opinion, however

the same alternator running at 6krpm and 42volts will make about 4kwatts while consuming the same 4.5kwatts input power
4/4.5=88.9% efficiency, which is fantastic in my opinion

(now my numbers might be off just a bit, because of graph interpretation of values, but you get the idea there is a dramatic increase
in efficiency)

it should also become clear that allowing the same alternator to run at even higher voltages, such as 57.6 for a 48vdc nominal battery
would run at even higher efficiency.

the reality in real life is a bit different that illustrated, the spread using a similar alternator to that which you are using from leece neville
will start at about 3kwatt at 14 volts and return around 45-50% efficiency, and if allowed to run at 57.6 for a 48volt bank will go well
over 7kwatts output and well into the low to mid 80's for efficiency.

in my opinion your application should actually be in the mid to upper 80's if it is matched well with the transformers, with the alternator
being allowed to run ~120vac feeding the toroid transformers to do the step down,

the toroids should get you into the 92% range i would expect, so

my guess would be a mid 80s' lets call it 85% alternator and 92% transformer pack for an adjusted efficiency of about
.85 x .92% = 78% efficiency

so in the end a move up from 45-50% efficiency to perhaps 78% is a dramatic improvement in efficiency!

and that should be the results of all your work.

bob g

[Lloyd]

bob g

in my opinion your application should actually be in the mid to upper 80's if it is matched well with the transformers, with the alternator
being allowed to run ~120vac feeding the toroid transformers to do the step down,

the toroids should get you into the 92% range i would expect, so

my guess would be a mid 80s' lets call it 85% alternator and 92% transformer pack for an adjusted efficiency of about
.85 x .92% = 78% efficiency

so in the end a move up from 45-50% efficiency to perhaps 78% is a dramatic improvement in efficiency!

and that should be the results of all your work.

bob g

Dear Professor,

If and when this project meets the pavement, my end results will be almost 100% reflection of your work (well that is if i don't f*6K it up). All I did is pose a question, which YOU stepped up and inspired me to proceed.

THANK YOU...no matter the outcome.

Lloyd

[Lloyd]

A method of calculating torque of a Lundell-type synchronous generator is disclosed
http://osdir.com/patents/Electricity-generator/Method-calculating-power-generating-torque-synchronous-generator-07375500.html

I like tidbits

& some meat & potatos


lloyd

#
BACKGROUND OF THE INVENTION
# 1. Technical Field of the Invention
# The present invention relates to a method of calculating power generating torque of a synchronous generator, especially, a vehicular Lundell-type synchrounous generator.
# 2. Description of the Related Art
# In recent years, attempts have heretofore been made to reduce the amount of fuel, such as gasoline, to be injected into an engine with a view to achieving the improvement in fuel consumption. To this end, the engine is controlled in a pinpoint precision to rotate in stability. In the meanwhile, an electric energy needed for the engine for safety and comfort purposes is rising year to year and a generator (alternator), connected to the engine, has been increasing in size. This results in an issue with the occurrence of an increase in torque of the generator accompanied by instable rotation of the engine. For instance, under circumstances where large electric load is turned on to rapidly increase the electric energy of the generator, a drop occurs in a rotational speed of the engine with a consequence of engine stop.
# To address such an issue, research and development work has heretofore been taken to provide technique of conducting integrated management of electrical loads through the use of a communication system inside the vehicle for predicting power generating torque to control the amount of fuel to be injected to the engine with a view to achieving stability in engine rotation.
# However, even if the related art practice has enabled the prediction of the amount of generated electric power (electric current in generated electric power), a difficulty has been encountered in accurately calculating power generating torque on this occasion. Power generation torque represents a turning force equivalent to energy to be consumed. In this case, energy to be consumed is a sum of generated electric power and the amount of loss dissipated in heat. For example, even with output currents at the same value, variation takes place in a coil resistance value depending on surrounding temperatures with the resultant change in heating value. Thus, a need arises for accurately obtaining the amount of losses of the generator depending on an operating status of the generator and usage environment thereof, otherwise no success can be expected in calculating power generating torque.
# Further, the amount of losses of the generator is broken down into copper loss, rectification loss, excitation loss, iron loss and mechanical loss. Among these, iron loss particularly consumed in an iron core results from hysteresis loss and eddy-current loss in mixture and, hence, a calculation formula becomes complex with the resultant difficulty in identifying the cause of losses. This results in deterioration in precision of calculation results. Moreover, a remarkable increase occurs in a time interval needed for calculation and it has been conceived that no computation of such factors is possible with a computer Installed on the vehicle.
# Therefore, attempts have taken to preliminarily assume the environment for the generator to be used and prepare a map covering a whole range of conceivable combinations of factors while actually measuring parameters such as temperatures or the like and retrieving target torque from the map. However, this results in a need for preparing torque data in a multidimensional approach in conformity to various statuses and, thus, considerable efforts are needed for measurements. This causes a remarkable increase in a volume of data with the resultant increase in a retrieving time interval and another issue arises in a difficulty of processing these data in an ECU. Moreover, a large number of sensors are needed for measuring input values of the parameters and cause an increase in production cost. Also, if the volume of data and the amount of inputs are restricted, then, another issue arises with a difficulty in obtaining high precision.
#
SUMMARY OF THE INVENTION

# The method of calculating electric power generating torque was established only by inputting the rotational speed N of the generator and current IB in generated power in a manner expressed below.
# First, excitation current of a rotor is computed. Excitation current of the rotor varies in substantially proportion to output current IB.
# During operation of the generator with the maximum current Ifull for each rotation, since excitation current takes the maximum excitation current Iffull current can be computed based on the proportionate relationship in an equation expressed as
# If = If full · ( I B I full ) . ( 1 )
# Next, gross current Igross is computed in an equation expressed as.
# I gross = kI B + If = ( 1 + If full I full ) ⁢ I B . ( 2 )
# Subsequently, DC/AC conversion is executed.
# In general, a rate k for conversion from IDC to an alternating current root-mean-square value IAC is expressed by a relational expression as IDC=kIAC  (3).
# The conversion rate k is probable to be slightly deviated due to distortion of a waveform and expressed in a theoretical figure of k=1.35.
# Current Ist flowing through the stator corresponds to a gross alternating current value and is expressed as
# I st = 1 k ⁢ ( 1 + If full I full ) ⁢ I B . ( 4 )
# Hereunder, respective losses are calculated in sequence. Since a total sum of losses is finally obtained, no order of calculating losses has an adverse affect.
# (a) Copper Loss:
# When using the alternating current root-mean-square value of the stator, copper loss becomes a total sum of heat generated for respective phases and expressed as Wst=mIst2R  (5) (WAC: copper loss of a stator R: resistance per phase m: number of phases). Resistance of the stator varies depending on temperatures and, so, a need arises for predicting the temperature depending on the magnitude of output current. Calculation is executed this time for each rotational speed based on a predicted value expressed as
# T = 140 ⁢ ( I R I full ) 2 + T o ( 6 ) (Wfull: full output for each rotation IB: output current to be calculated).
# Upon estimating such a temperature, a resistance value can be estimated as
# R = ρ + T ρ + T o ⁢ R o ( 7 ) (To: initial temperature, Ro: initial resistance, ρ: coefficient), where ρ represents an individual value depending on a temperature of each metal and is expressed as the coefficient of copper in the present embodiment as ρ=234.5.
# Substituting Equation (6) to Equation (7) to newly organize the coefficient gives R=(k1+k2IB2)RO  (.
# Substituting this Equation ( to Equation (5) to reorganize Equations (1) to (7) gives
# W st = m ⁢ { 1 k ⁢ ( 1 + If full I full ) } 2 ⁢ I B 2 ⁡ ( k 1 + k 2 ⁢ I B 2 ) ⁢ R 0 . ( 9 )
# Further, reorganizing this Equation with the coefficient gives Wst=αIB2(k1+kIB2)  (10). Thus, calculation can be executed in computation using a square value of IB.
# (b) Rectification Loss:
# Output current to be converted to DC current flows two times through an earth-site and a B-site to take an average value of IB and can be computed in a manner expressed as
# W rec = 2 ⁢ Vdi ⁡ ( I B + If ) = 2 ⁢ Vdi ⁡ ( 1 + If full I full ) ⁢ I B . ( 11 )
# (c) Excitation Loss:
# Excitation current is requisite and a product of excitation current and voltage results in excitation loss. As excitation current is small, excitation voltage is also small and, after all, excitation loss is assumed by an Equation expressed as
# W rot = V f ⁢ If - V D ⁢ ⁢ C ⁡ ( I B I full ) · If full ⁡ ( I B I full ) = V D ⁢ ⁢ C ⁢ If full ⁡ ( I B I full ) 2 . ( 12 )
# (d) Mechanical Loss:
# Mechanical loss can be approximated only through power generation frequency N and, hence, is obtained in polynomial Equation expressed below. Here, a 2nd-term can be expressed as =0 in conformity to the degree of a need in precision of a torque calculation value. Wmech={γ·f+ε·f2}  (13), where γ and ε represent coefficients and f represents a rotational speed.
# (e) Iron Loss:
# From experiment data in the past, it has been predicted that with the Lundell-type synchronous generator, iron loss is proportionate to an output current value. This is predicted because a synthesized magnetic flux is distorted due to reacting magnetic flux resulting from output current and an iron core of a Lundell-type rotor is made of core formed in a non-laminated structure in general practice to be apt to cause the occurrence of eddy current.
# A basic principle of iron loss (loss in an iron core) occurring at a maximal current Ifull for each rotation of the rotor is supposed to have a correlation with hysteresis loss and eddy current loss and predicted to take functions of a first power and a square of a frequency. Given the above, Equation (14) is selected.
# Using a regression curve resulting from the experimental results determined α and β.
# W core = { α · f + β · f 2 } ⁢ I B I full ( 14 ) (α, β: coefficients, f: a frequency in power generation, IB: output current).
# With the present embodiment, α, β were set as α=63 and β=3. Based on such setting, computation was conducted. Here, iron loss is compared to that of the related art practice.
# In the related art practice, a calculation formula of iron loss is generally expressed as Wcore=B2{σHf+σεd2f2}U  (15) (σH: coefficient of hysteresis loss, σE: coefficient of eddy current, d: plate thickness of a sheet, U: a total weight of iron).
# Further, it is a general practice to alter coefficients of a thickness portion of a stator core on a rear side thereof and a teeth portion in the same Equation for calculation.
# The magnetic flux density B varies depending on a rotational speed, a voltage and an output current value.
# In particular, an inner voltage Equation, involving a component corresponding to a drop, is expressed as Vi=√{square root over ((VO+RIAC)2+(XIAC)2)}{square root over ((VO+RIAC)2+(XIAC)2)}  (16).
# Converting this inner voltage to the amount of magnetic fluxes gives
# ϕ l = V l 2.22 ⁢ P 2 ⁢ T 120 ⁢ ⁢ rpm × 10 - 3 . ( 17 )
# Dividing the amount of magnetic fluxes by a magnetic flux path cross-sectional area S results in the magnetic flux density as expressed
# B = ϕ l S . ( 18 )
# Substituting the magnetic flux density B to Equation (15) allowed iron loss to be obtained.
# The calculation needs to execute large numbers of steps using complicated equations as required in the calculation stated above. Also, such a calculation has a low precision with the resultant power generating torque in an increased discrepancy from a correct value.

[Lloyd]

yumm meat

http://www.patentstorm.us/patents/7375500/fulltext.html
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
# An engine, shown in FIG. 1, drives a vehicular generator (alternator) via a belt. The generator (alternator) 1 generates an output converted to a DC power and then supplied to a battery and electrical loads of a vehicle. The generator 1 is controlled to provide a power output with an output voltage at an appropriate value in response to a detected voltage of the battery to which the electrical loads are connected in normal practice. A computation device 2 is placed to compute information indicating a torque for power generation (power generating toque) from information given from the generator 1.
# An engine controller ECU 3 retrieves, from the computation device 2, information indicative of a power generating torque produced by the generator 1 depending on the need and determines the amount of fuel to be injected to the engine at a rate controlled in view of the power generating torque information, thereby achieving stability in the rotation of the engine.
# Next, a method of calculating a torque is described below. An IG (ignition) terminal of a regulator (not shown) in the generator 1 detects a vehicular key switch being turned on. Then, a torque calculation program is progressively executed. As shown in FIG. 1, the computation device 2 is supplied with an output current value IB and a rotational speed N from the generator 2. A rotary sensor (not shown) directly retrieves the rotational speed.
# The computation device 2 executes the computation based on such inputs to provide power generating torque information as an output.
# FIG. 2 shows a flowchart of a basic sequence of executing torque calculation. Output current and rotational speed are taken and using these information allowed calculations on an output (W) and loss (W), upon which the computation is executed to obtain a demanded drive power (W) equivalent to a sum of the output (W) and loss (W). Since the rotational speed is input, the computation device 2 can calculates torque based on a value of the rotational speed and drive power to provide a calculation result as an output.
# Hereunder, a detail description is given of a method of calculating the amount of loss with reference to FIG. 2.
# First, upon receipt of calculation command signal, the operation is executed to generate electric power while a flow of the present calculation starts. The current sensor provides the output current IB and the rotary sensor provides the rotational speed N. Also, the rotational speed may be extracted based on an output frequency of an alternating current voltage of the generator.
# In step 102, the operation is executed to compute maximal current Ifull of product information based on a value of the rotational speed N using an approximation formula. This value may be extracted from a map. Also, maximal excitation current Ifull is preliminarily set as product information.
# First, in step 103, the operation is executed to compute excitation current of a rotor.
# Excitation current If of the rotor is obtained by Equation (12) as expressed
# If = If full · ( I B I full ) .
# In the next step 104, gross current Igross is computed in a formula expressed as
# I gross = I D ⁢ ⁢ C + If = ( 1 + If full I full ) ⁢ I B
# Subsequently, in step 105, AC conversion is executed.
# In step 106, copper loss is calculated based on Equation (19) expressed as Wsi=αIB2(k1+k2IB2)
# In succeeding step 107, rectification loss is calculated using Equation (17) as
# W rec = 2 ⁢ Vdi ⁡ ( 1 + If full I full ) ⁢ I B .
# In the next step 108, excitation loss is computed in a formula expressed as
# W rot = V D ⁢ ⁢ C ⁢ If full ⁡ ( I B I full ) 2 .
# In step 109, mechanical loss is calculated in a formula expressed as Wmech={γ·f+ε·f2}.
# In succeeding step 110, iron loss is calculated based on Equation (19) as expressed as
# W core = { α · f + β · f 2 } ⁢ I B I full
# In step 111, output power is obtained in a formula expressed as Wout=VBIB.
# In step 112, a required input (W) is obtained as a total sum of the factors expressed as Win=Wout+Wst+Wrec+Wrot+Wmech+Wcore.
# In step 113, the above factors are converted to Tgen=(Wout+Wst+Wrec+Wrot+Wmech+Wcore)/(2πN/60).
# In step 114, a calculation result is output.
# Thereafter, the operation stands ready until a subsequent calculation command appears in step 115 and upon receipt of the command, the above steps are repeatedly executed again.

[mobile_bob]

Lloyd:

you keep trying to stuff this crap into your head and it is going to explode!

(in this case "crap" does not mean useless info but rather complex and perhaps inapplicable info)

i say this because i am often ridiculed for making things overly complex and am no fan of KISS at all cost.

here is my point, in an earlier post you posted about SMR, switch mode rectifier or controlled rectification.
i have long been a proponent of such for use with windgenerators, going back about 10 years now, only recently
has there been interest among the windpower boys, however they for the most part although seeing the value
lose interest because of the added complexity of the electronics.

they really could benefit from such technology because of the variable nature of their power source, and the need of the
blade set to run in a rather narrow speed range in order to reap maximum efficiency.

this same theory of operation is being applied with computer controlled alternators in the automotive industry because of the
variable speed nature of the power source and the desire to maximize efficiency of the alternator over that broad rpm range.

basically the smr system works like a transmission with an infinite number of gears, allowing for a matching of power to load
over all anticipated operating parameters.

it is my belief that we really don't need this technology, because

we are running at a fixed speed (although you might be different because of the variable speed nature of your prime mover? or are you driving the alternator with a dedicated engine? i have long since forgotten the specifics in your case)

we can optimize the alternator to produce the requisite or desired power, at a fixed speed, and use the balmar to finish of the load matching to the batteries (which it does remarkably well with the "amp manager")

the next test for you that would likely give you all the info needed for the transformer engineers, is to load test the 3phase AC output of that stator at the speed you want to run it at, at about 6amps field current (i mention 6amps because i am relatively certain your rotor can handle 7amps max, so there will be some margin of safty)

i would apply a 3phase AC load onto the alternator and see what the output voltage drops to at that field current that provides about 30amps from the alternator,  i am thinking that maybe at 6krpm and 6amps field at about 120vac phase to phase you should get 30amps without much problem.  if so then a 10:1 step down transformer pack of suitable power capacity should net you over the 250amps you are looking for.
although i might still go for 8 or 9:1 step down so i know i have enough headspace to make the required voltage for charging, giving up a bit of amperage which should still leave you close to 220amps or better.

now for a load bank, let me know if you would like to come down to tacoma, there is a salvage yard down here that has a large batch of wire wound power resistors that i think would work well for testing, and they sell for about 2 bucks a lb or so.

three of those would make up a load bank that would work well for your application i would think.

bob g

[Lloyd]

Bob,

I just started my research on the SMR...so don't have any conclusions...My initial thought was that it may bypass the need of the step-down transformer...I don't know enough to make that decision.

But I just got off the phone with Osbourn Transformer, we spent about 45 min. and have specked out a proof of concept transformer, which will include 3 taps on both sides per phase to allow for some margin of tinkering during testing.

I will be running at fixed speed, during bulk charge cycle. Also Glen at KEI said I could put a max of 18 volts on the rotor all day, so that works out to about 9 amps at 2 ohm rotor cold.

My numbers are based on 5.5 amps cold, Osborn suggested that we work on 7.5 amps to give some over head.

They will start winding the transformer/rectifier tomorrow...but I have one more task, they want the  ohms per foot, and the inductance of the conductor cable for the high-voltage side, to run losses prior to transformer. So I have to get a hold of American Cable to get those numbers, this morning.

Bob, you bet I'm almost ready to come to Tacoma...as soon as the transformer/rectifier lands in Seattle, which will be by the end of next week.

Thanks again professor, if you want to get your testing hat ready.

Lloyd

[Lloyd]

Just off the phone with Hurley Wire Co...

I now have sourced the three phase cable...it will be 6-4 shielded 400hz SO cable. It is commonly known as jet starter cable.

Osbourn was able to get the ohms per 1k foot, along with the inductance, and impedance.

Now they are ready to wind the trans/rec.

Lloyd

[mobile_bob]

how much do they want per foot for that cable?

bet they don't give it away!

but i guess all copper cable is expensive anymore, regardless of the intended use.

bob g

[Lloyd]

bob,

This is  a link to the 3phase 400hz cable https://edeskv2.belden.com/Products/index.cfm?event=showproductdetail&partid=1650

As it turns out the big difference is the shielding, a look at the NEC show that do to the skin effects a cabe to be properly sized in amplitude goes next up...so what i originally planned was a 8-4 cable has to be a 6-4...I might still be able to use a 8-4 but the amp potential say I need  #6, even though with the losses and heat tha alt isn't going to produce it's paper potential.

O well I'm use to up sizing  cable the fractional cost is marginal at best to maintain a good V-drop %

Lloyd

[Lloyd]

PentodePress has resistance, inductance, and reactance: http://pentodepress.com/calculator/wire-inductance.html

From http://www.prodigy-pro.com/diy/index.php?topic=31900.0 which has a great topic discussing resistance, inductance, and reactance.

Lloyd

this is just a link to some automation control parts http://raise.rockwellautomation.com/catalog/_0.asp?onKil=http://www.ab.com/raise

and this link to the military service education publications http://www.google.com/custom?domains=www.tpub.com&q=alternator&sa=Search&sitesearch=www.tpub.com&client=pub-8029680191306394&forid=1&ie=ISO-8859-1&oe=ISO-8859-1&safe=active&cof=GALT%3A%23008000%3BGL%3A1%3BDIV%3A%23336699%3BVLC%3A663399%3BAH%3Acenter%3BBGC%3AFFFFFF%3BLBGC%3AFFFFFF%3BALC%3A0000FF%3BLC%3A0000FF%3BT%3A000000%3BGFNT%3A0000FF%3BGIMP%3A0000FF%3BLH%3A50%3BLW%3A700%3BL%3Ahttp%3A%2F%2Fwww.tpub-products.com%2Findex.2.jpg%3BS%3Ahttp%3A%2F%2Fwww.tpub.com%2F%3BFORID%3A1%3B&hl=en


3phase basics alts/trans http://www.elighttraining.com/files/ag4.htm