 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Other Stuff : GE 5SME22NJ028 Micro-hydro
| Author | Message | ||||
| davef Guru  Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Herb, The main reason I was considering this motor was the desire to get as much efficiency as possible at the 300-600Watt input level. I have EcoInnovation's manual of the F&P series and also am aware of a cut down SmartDrive unit to try and get the efficiency up. Some facts: I have about 35-40metres of static head, 300 metres of 50mm pipe. 1L/s with the option of going to 2L/s during the winter. This should give me about 150-300Watts at 50% efficiency. You said this unit got warm at 24Volts and 6 amps. Were you using all the windings? I was hoping it would work up to 600Watts input. Perhaps the normal F&P unit is required. During the summer I would only be taking 1/2 to 1L/s and want good efficiency. I would probably buy a rotor and buckets from EcoInnovation rather than trying to make them. Voltage, probably 24Volts even 48Volts if it makes sense. I plan to have the batteries at the machine end with a 240Volt inverter feeding power up to the house about 50 metres away. My neighbour has a full SmartDrive unit cranking out 700Watts, or maybe I should say had as it seems to be a regular job to replace bearings. He was looking at a Platypus unit. Any general direction you can provide would be appreciated. I am still at the "information gathering stage" as the house won't be started for another 6-9 months. Thanks, Dave |
||||
GWatPE Senior Member  Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I would not be happy with 50% efficiency. Would be useful to compare output from an AxFx unit. Is most of the efficiency lost in the pelton wheel, or the alternator? Gordon. become more energy aware |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Gordon, It is a figure quoted in several pieces of literature. I assumed it included both the pelton wheel and generator, but to what extent? Pelton wheel efficiency Seems to indicate that a properly designed and run Pelton wheel has high efficiency . . . 80-90%? Therefore, I have to conclude that the generator is responsible for the lion's share of the "50%" figure. Would be interested in what efficiencies you get from various generator configurations. Dave |
||||
oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Wired for 240 rpm, I will get 500w@ 86% efficient.(.7R) Wired for 1000rpm, it would be almost only bearing loss as resistance could come down to .17R or less. for 50v.... So e=ir=10A x .17R= 1.7v loss. Watts lost in stator will be esq/r=17W losses..... or around 95% efficient??plus maybe bearings. In that configuration 1000w input(20A) would be 93%... 3.4v loss so 3.4x3.4/.17=11.56/.17=68watts lost in the stator..so 932w out down to 93% At 5000w input, were down to 66%.. and toasting the stator with 1700 watts of heat and 3300 output. The figures are better than that, as most likely V=55v for the 48v pack. This will lower amps and so lower losses again. With axfx, you need strong magnets, but have no iron losses, and can get very low resistance windings because of that flux. Add your high rpms (compared to wind), and it looks very good. With .17r stator, you may need the batteries up at the house, and use the cable run as a resistor, or you may stall the pelton altogether.... I don't know hydro. Herb, your 60W bearing loss I assume you mean bearings and iron loss? 60W in the bearings would be pretty toasty bearings and plastic supports. .........oztules Village idiot...or... just another hack out of his depth |
||||
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi Dave If you are ever in the top of the south island arrange a visit. your figures show 100% eff 387watts a pipe loss 1.5 metres at 1L/sec . most consider 50% eff to be guide line this is overal,it is right large units get up to 90% but not on small units. In fact when consulting I always state 30% so no disapointment occurs we do miss out a few confirmations but least no come backs later. The highest I have achieved is 60%. ie 10A out of 110 Metres .2 L/sec on a gentle annie running 2800rpm. To get 1200rpm you wheel will need to be 208mm pitch dia To get 3000rpm 83mm this would be best for GE motor however Micheals cups are two large for this . your jet size for 1 l/sec is 7.4mm so his cups ok there. I would be recommending a smartdrive unless you can arrange smaller cups I make my own but only for self use. In its standard from I only use the low voltage winding on the GE , however there is great potential to use the high voltage winding over 400volts and current low for same or more power here and built a dc to dc converter. I have in the past used three Phase transformers but due to highish frequencies and harmonics, losses in cabling and transformers is unacceptable. Eco units do do bearings in due to water entry, I have one here at the moment for repair . Units I build of similar design have bearings that have been running 7 years in one case. Will pass on info if you are building a unit. Platypus use induction motors excited by caps from all the ones I have seen not very efficient, but well built. To run our house with all mod cons fridge, freezer,dishwasher,microwave, jug, toaster etc we have no real savings needed if we produce about 400watts less than that we start saving jug toast dishwasher locked away etc (cooking is gas ) Oztules You a probably right about your eff but I would myself only see bench testing with accurate input measurements compared to output, has this been done on Otherpower ? I will have to confirm 60watt figure tested long time ago but seem to remember bearings only, also Dr chalko also measured this was about only figure I agreed with. Just measured temp on my main unit inner housing 17 cent outer (away from water cooling shaft) 22 c The eco unit I have been running on test lower but water is leaking onto the housing so giving direct cooling. note housing 10 c so there is a fair bit of heat. Herb |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Herb, Thanks for the invite. I noticed that one can not PM(personal mail) on this forum . . . well I can't find it. My email is dave_festing(at)hotmail.com According to Michaels' info running a 100S at 48V in Star should be good for about 75% from 100 to 300 (400) Watts. A 100S at 48V in Delta >75% from 100 to 600 Watts. Would prefer 24Volts, so a 100S at 24V in Delta shows about 70% from 100 to 300Watts. Those are output powers. Now, using all the windings on the 5SME22NJ028 and sending 3 phase power to the house is sounding more attractive. Rectify at the house and build a buck converter down to 28V. What voltage would you expect out of an unmodified SmartDrive (model?) for the 1L/s example? 200-400Volts RMS per phase? Edit: I have found a graph on page 71 in Michaels' manual that gives an idea of the open circuit voltages. The bigger wire size and fewer number of turns give the largest open circuit voltage. There must be wiring differences between the 100S, 80S and 60S or stronger magnets or ? Likewise, if you are coming down to Christchurch drop us an email. And thanks for the info! Regards, Dave |
||||
| herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Dave The highest voltage open circuit would be 60s all magnets same 100s has 44 turns / pole 80s has about 100 60s about 150 (memory not so good on last two ). I would have to check what volts are now not good keeping records. I have long ago given up reconnecting the smart drive I rip all windings off and put the turns & wire size for the job. ie the equivalent of a 100sp is only 88/14 = 6 turns / pole a easy task while watching tv in facr a job the wife usually does. at 1200 rpm this will pump 30 + amps into 24v batteries. the turns /phase = 88 here if we had a unmodified 100s turns / phase is 44 * 14 = 616 so you can see open voltage going to be high note 80s 100 * 14 = 1400 60s 150 * 14 = 2100 I myself would not run three phase AC the high frequencies due to the fundemental andthe harmonics caused by the step fronted pulses charging the batteries only at the peak cause big loses in the reactance values of the cable not long ago I ran a test with a cable here and responded to a thread here but can not find it now. I always have rectifiers as close to windings as possible, use your DC dc conveter at the batteries. I have always wanted to build one and have been watching Oztules and Dingles talk here on converting PC supplies with this in mind Herb |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Herb, Your machine type versus turns and the resultant voltage makes sense. Must have got the wrong end of the stick from his graphs. I know toroidal transformers are sometimes used, but was going to actually test two of them back-to-back to see what their frequency response is really like. I take your point about loss in rectified power if you are not passing all the harmonics. I will have to carefully consider the pros and cons of having all the plant about 70 metres away from the house. Thanks again for the tips. Dave |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I had a 300W 50Hz torroid, factory rewound for operation at 400Hz. The same iron core passed 2.5kW. The wire turns and section on primary and secondary reflected the increased power. The frequency of a 1000rpm F&P is over 400Hz, so this would make an interesting rewiring challenge for a torroid makeover. Better to rectify and run DC. If the alternator has AC voltage multipliers then most of the loading and power tracking problems disappear. A simple solar type MPPT can be used for DC-DC conversion. All that would be needed is overvoltage protection on the alternator output. This could be a simple closed loop type regulating input DC-DC buck conveter. Early, pre microprocessor solar switching converters use this arrangement. Anyone with a high voltage system requirement could easily test this. Gordon. become more energy aware |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Gordon, "The same iron core passed 2.5KW". Are you saying because the core was much more efficient at 400Hz that the power lost was the same as the original transformer run at 50Hz loaded to 300W? Did it require lots more turns? My brief look into buck converters seem to suggest that you don't want to drop the voltage by more than 1/2 to 1/3 of the input. For example, output voltage 13V input voltage doesn't want to be greater than 39V. In order to keep efficiency high. Herb, Your comment about cables reactances modifying the waveform. Transmission line theory . . . 1kHz has a wavelength of about 300KM in air, maybe about 200KM in plastic based transmission lines. For any modification of the harmonic content, say 3rd or 5th harmonic the length of the "transmission line" would need to some appreciable length, my guess 1/16th to 1/8th wavelength at 5KHz for example. So, the 5th harmonic or 5kHz has a wavelength of about 200KM/5 = 40KM. 1/16 of 40KM is about 3KM. My transmission line is about 3000M/70M = 2.3% of my guessimate. What got me thinking about this was if the batteries are at the hydro plant end and I use a modified sine wave inverter to send 230VAC up the transmission line, what effect would that line have on the non-sinusoidal waveform produced by the inverter? Maybe if would only improve it as far as running transformer operated equipment on it. If you want to give me some figures, source impedance, load impedance, harmonic content, wire size, length and spacing I can throw it into a circuit simulator and see what comes out. I can see source waveform modification occurring when a relatively "long" line (ie 3-6KM) is not terminated in its characteristic impedance. If the line had an impedance of 100 Ohms (at the fundamental and all significant harmonics) and it was not loaded with 100 Ohms, ie about 500Watts (P = (230*230)/100) then you are going to get reflections back to the source. Dave |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Dave, I did not say this. The iron core losses in a torroid are apparently not major loss. The wire seems to be. The factory rewound torroid had approx 1/8 the turns on primary and secondary, but the wire Xsection was approx 8X. The computer power supply could be a better choice of voltage converter for 12V. 12V is still a poor choice of system voltage for a system above 600W. Gordon. become more energy aware |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
OK. So, core efficiency didn't seem to change much, it was only the wire loss that decreased (due to the larger wire size). Before I go and buy $150 worth of transformers and rewind them with 1/8 the turns and 8 times the cross-sectional area wire . . . will I be able to put 8 times the power through the transformers before I reach the same temperature rise as when they were operated at 50Hz at their original power rating. If so, then I could buy quite small 50Hz toroidal transformers, for my 300W system. I'll do some research on computer power supplies. Thanks for the information. Dave |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Herb, I spend most of my life designing RF stuff, so perhaps the transmission line stuff is not relevant. Is it just the capacitance of the transmission line that can upset the performance of a F&P? I looked at the output of the subject device and it looks like a pretty good sine wave. Cheers, Dave |
||||
| herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Hi Dave Yea I dont see RF theroy here. I have never investigated all the reasons power is reduced but capacitance would be my main suspect. I here have basically two systems the shed power with local solar , wind and hydro when these batteries get charged it dumps to the house batteries 100M away using the modified waveform inverter and 2.5mm TPS and transformer charger. never measured losses but would not be great. I think sometimes the KISS principle overrides efficiency sometimes. I also have a friend/client that has a hydro with a 1KM run to the house this originally we ran on dc low voltage as his demands were not great. but it is now converted to having batteries and a cheap inverter and charger it gave a signifigant increase. ( little note here before I was involved he approached a firm that advertised the use of torrid transformers, but when asked to quote by him stated that they were not advisable and quoted big cables at 48 v ) You are right to investigate this before building I find that physically doing every thing in building setting up here very little time to play. The system tends to evolve rather than be planned as it grows to met demands. Herb ps both FP motors give a respectable looking sinewave,when off load and supplying resistive loads,must measure the form factor one day |
||||
| davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
Herb, Thanks for added info. The wind and the hydro sources are both about 70-100 metres from the house. The solar will be on the roof of the house. Seems best to have high voltage AC from the wind and solar to cut down feedline losses. I had a long chat with our power supply expert and he said, yes buck converters are really only good for a 2-3 times reduction in DC voltage. Maybe that forces one to a 48Volt system. The other option is to use flyback converters which allow any ratio voltage drop or boost, but he said they are very noisy (RF-wise). Whether this would be a problem in my situation, has yet to be determined. Dave |
||||
| herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258 |
Dave seems to me that all old pc supplies are suitable or at least supply a source of parts. All use a transformer so can be adjusted by turns ratio classified as half bridge I think. The loses may be high with a converter but remember that the rectifying and transporting at high voltage will save heaps. Problem I have struck with 48v systems is the lack of reasonably priced inverters. Manufactures seem to think that if you are at this level you need all the bells and whistles, may have changed nowdays. Herb |
||||
 Print this page |
