Anotherpower.com Forum
Renewable Energy Questions/Discussion => Automation, Controls, Inverters, MPPT, etc => Topic started by: Dr_Zogg on August 17, 2015, 04:45:01 am
-
howdey all,
i have two powerjack inverters here that i am hoping to mod for my off grid setup.
the first inverter is a 6000w with one transformer and a fan.
the second one is a 15000w with 3 transformers and 3 fans.
i am on the hunt for 2 or 3 aerosharp or inspire inverters to do a double or triple? stack for the 15000w.
i have got and stripped down an aerosharp 3.5 with the hope of rewinding just that one single transformer to go into the 6000w?
if that would be a good option would you have some advice on how to go about it?
i have read oztule's other posts a fair few times to try to familiarize myself with it before ashing questions
the pics show the transformer that i have stripped from the aerosharp 3.5kw it weighs in at just over 22kg
thanks in advance for the help.
-
I have a few of those 3kw aerosharps ( haven't used any of them yet in this application)
If you remove the outer winding (250v I think) you will be left with a 230v winding ready to use.....
I would have no problem using that for the 6kw unit... non at all... will do it easily, as there will not be 6kw on for very long at all, and will run 4kw all day without any real cooling.... a winner.... and even better if you were to wind another 20 turn on it to bring it up to 250... this will lower the idle current a bit more as we move the saturation point further away.
Not as good as winding your own, but very fast. The fact it is not potted in any way, will make it a bit noisy when under full loads is the only difference... and that the life may be compromised like all other commercial transformers of torroid design.. they never pot them... so not a real concern either... just an observation.
So check the voltage on the inner winding... is it 230 or 250v ( hopefuly 250v but I think 230v)
If you use the same box, it will look similar to this ( this is an 8000w board set in that box)
[attachimg=1]
...............oztules
-
thanks for the quick reply.
sorry to sound like a dumbass, but how do i go about checking the voltage of the inner winding?
Cheers,
Sam
-
Look at the label... colour code is there... look where the wires go.
You have to remove the outer winding anyway, so easy to follow the wire colour there..... and see which one your removing.
I think it is 4 in hand 1.6mm diameter for that winding. I unwound one of them a good while back.... but the core is still on the shelf.... and only vauge memory remains. As you can see I used the box, but not that core.
Let us know when you know.
................oztules
-
thanks heaps for that. i must have been having a senior moment haha.
well I'm going to have some dinner and then get started, wish me luck.
i will put up an update and some pics later on.
what size wire and how many turns would you recommend for the primary?
also for the e core is it possible to use one of the rings i have kicking around from pulling these inverters apart instead?
Cheers,
Sam
-
sorry bout the delay. i got a new phone and it wouldn't play nice with my laptop >:(
well it is all stripped down and damn. you weren't kidding about getting sore shoulders.
i have decided to take your advice and add 20 turns to it. ( if i have gone this far i may as well do it right)
also it will be good practice for the next one.
i presume that i add 20 turns 3 in hand? also as for my questions in the previous post i have a heap of 4 gauge car audio
amplifier cable is this suitable?
now to go hand straighten this wire ;D
thanks for the help.
Cheers,
Sam
-
Sorry I missed that...yes the extra turns the same as the rest of them... 3 in hand.
Fun undoing it isn't it..... use the three bearings to straighten the wire.
For 48v systems:
Audio cable is about 20mmsq.. so if your going to run this thing hard, then two in hand, if you don't expect longish periods over 1- 2 kw, then one in hand will squeak through.
For the primary we want to know the volts/turn... did you happen to get the turns number when you unwound?
If not..... I am guessing it is about 1.1v/turn, so you will want about 30 turns on the primary for 48v... 15 for 24v and definately 2 in hand and more for lower voltages, and more for higher powers too.
To make certain, you can connect 240v into the 230v winding, and wind a few turns on to get the volts/turn on the new turns. You will have to soften the startup as the fuses will likely blow. You can use a globe and then bypass it when operating..... check the voltage on the 230v winding, and the turns on the test winding.... just to make sure 240v is available or thereabouts.
When you get the turns/volt or volts/turn in this case, divide it into 33v to get the turns you want... for 48v system.......... ie 33/1.1v=approx 30 turns ......scale to match other voltages.
It is a physical thing working on these transformers for sure.
keep at it.
.....................oztules
-
i finally got off of my lazy backside and did some work.
i have straightened the wire and rolled it onto a large cable spool 3 in hand. my hands knew about it after doing that haha
i built a bearing straightener like yours but ended up not having to use it, after wrapping the end of the wire around one of the stumps on my house and giving it a bit of a tug with the ute it ended up dead straight. is that normal?
anyway i have found some large welding cable for the primary. i will measure it later and see what you think of it, and all being well i am hoping to add the turns to the primary tonight.
Cheers and thanks for the help,
Sam
-
hi,
i think i understand the principle for checking the volts per turn that i add. but would it be possible for you to show me a sketch just so i can make sure i am thinking correctly? just because the way i am thinking of doing it i would make a giant 240v electromagnet? i just don't want to toast myself or the transformer :o.
thanks again,
Sam
-
i just measured the welding cable i have and it is 7.5mm across the core. do you think it would be suitable?
sorry for all the questions,
Cheers,
Sam
-
Welding cable is fine @ around 35mmsq .. I guess........ for 48v system for the 6000w converter, and even for the 8000w converter will be fine.
For 24v ok if not driven too hard for too long.
The turns per volt is simply running the transformer normally as is............. with the 240v winding to the 240v ac. ( may need a decent dropping resistor until it gets going..). then having a few turns on it and measuring the voltage for the say 10 ( or whatever you decide to average it out over ) turns you put on.... ie if you get 12v over 10 turns... then 1.2v/turn... I expect near that .. 1.1 to 1.2v/turn.
You could use the power jack for the 240v input.... just hook the output 240v to the 240v winding BEFORE you turn it on.... that way it will soft start the torroid. Hard start will blow your secondary fuses in the PJ for near certain.
Stretching the wire with a car is normally for the primary :)
..............oztules
-
oops. i thought i saw somewhere that you stretched the enameled wire. at least i didn't snap it then. worked well though haha.
i should mention that i am building a 48v system.
thanks for all your help Oz, you truly are a credit to the human race with the way you share your knowledge.
Cheers,
Sam
-
Check out my amazing graphic design skills!! ;D
does it look correct to you? if not. what have i messed up in this mess i call a brain? haha
Cheers,
Sam
-
......well... giving the enamel wire a bit of a tension up is one thing,,, using the ute to do it brings it to a new level :)
Graphic arts is better than mine.
That gives you the circuit, now wind a few turns around the torroid ( through the center ...) and measure the voltage that gets induced into these new coils.
This will give you the volts per turn in AC rms ( root mean square )... the equivalent of DC in power terms. ( the peak AC voltage will be 1.414 times the rms value... but we won't bother with this at this time). Use a normal multi meter to measure the volts and we will be fine and in AC rms.
I expect to see 1.1 - 1.2v /turn or in that region.
Once it is going, you will need to short out the globe to get the voltage up to 240v or thereabouts.
Proof your graphics are better than mine......
[attachimg=1]
.............
-
thanks heaps for that. i will put the turns on tonight and give it a crack tomorrow when i finish work.
nothing wrong with the artwork mate, get's the point across perfectly.
Cheers,
Sam
-
i did the tests tonight. and i got 24.5 volts from 20 turns. that works out to 1.225v per turn. i am assuming that equates to 40 turns for the primary? maybe i screwed up the math.
also, sorry to be a pain in the ass but would it be possible for you to give me a link for a suitable e core for the primary, I know i have seen it in one of your threads i just can't seem to find it.
Cheers,
Sam
-
25 turns will get you around the 30v mark, and this should be about where to aim for.
The ferrite choke ecores were E65. I had them, others may have had links........, so try ebay etc.
Glenn at the backshed has made one here too http://www.thebackshed.com/forum/forum_posts.asp?TID=7800&PN=1&TPN=1
They are turning up all over the place.
............oztules
-
25 - 26 turns will get you around the 30v mark, and this should be about where to aim for.
The ferrite choke ecores were E65. I had them, others may have had links........, so try ebay etc.
Glenn at the backshed has made one here too http://www.thebackshed.com/forum/forum_posts.asp?TID=7800&PN=1&TPN=1
Thats a single core unit as well.
They are turning up all over the place.
............oztules
-
So, now i have my extra 20 turns 3 in hand on and soldered to the existing winding's.
so i smack a couple of layers of mylar tape on before winding the primary?
then it will be onto the fan speed controller. i know i am working at a snails pace but i am also building the house at the same time.
that and work/family but i will get there in the end. well i hope i do ::)
Cheers,
Sam
-
Yes providing the 20 turns are the same direction as the original ones...... if unsure, you can inject 240v into the original winding and then measure 260v across the whole winding... if you see 220v... you wound it backwards ......
Then tape and primary.
...............oztules
-
Sweet, thanks for that. i will put up some more pics as i go.
what would you estimate this transformer to be capable of continuous and surge?
once again thanks so much for all the help.
Cheers,
Sam
-
bring your primary up to 26-28 now.
Transformer wattage is very much driven by the ability to cool the thing. In a closed box and sealed, that tranny was good for 3kw all day as a grid tie. That will still hold.
Keeping the magnetising current down will help keep temp down too, as eddy currents will be lower, core temp from magnetising the steel to lower testla will help also.
Making the thing 260v is what helped this... pushing the magnetising current down, ie.... the idle current.
6kw would be comfortable for shorter runs if the primary is large.... 50mm sq etc.
Transformer cores don't saturate from over current if they are not saturated at idle any way......contrary to popular belief.... only voltage and frequency can do that, so the R in the circuit will dictate the upper limits..... All the extra current in an overload goes into inducing current into the secondary... ie between the coils themselves, not saturating the core...the core field remains the same .... like in a motor where all the extra current finds it's way into driving the load, not saturating the core.... any heat build up then is due to Isq x R loses in the wire.
.................oztules
-
I am finally back at it :)
just a few pic to show where I am at with the mock up on the first inverter.
basically just have to make the fan controller now and then I should have everything I need
to assemble it.
I will keep the updates going as I work on it now. and no doubt the questions also haha
Cheers,
Sam
-
yesterday I picked up 5 inspire 1.5kw inverters brand new in sealed boxes for a good price.
so my plan is to use 2 of them as is connected as grid ties, use 2 more ad donor cores for my big transformer
and have one as a spare.
I do have a couple of questions though.
I am thinking of making the transformer for my larger unit oil cooled and was wondering.
(1) would the construction of the transformer remain the same or would changes be needed?
eg don't use resin to allow the oil to "soak" into the transformer?
(2) what do you think the approx capabilities of a single core unit like I have already made and a double core
unit like you have made would be increased to?
I would fold the enclosure up out of some 2mm aluminium sheet and tig it together/add cooling fins.
probably make it 2-3 times the size of the transformer so that it holds an ample amount of oil.
Thoughts?
Cheers,
Sam
-
Sam... my rule of thumb.. if your inverter was 1.5kw, then you can use 1.5kw continuous.... 6kw surges, 3kw 15min.
If you double stack and rewind ( reduces copper length by 25% )then that increases to 4kw continuous, and pretty much whatever you want in surge and 6kw for 15min.
Your cooling will dictate real world limits. I can't comment of liquid cooling except to say that the power companies use it because it makes the transformer size manageable.
..................oztules
-
thanks for that. I might build a spare one and have a go at oil cooling it then push it hard for some testing.
I am just trying to build a solid separate unit to power the shed I have some pretty heavy duty equipment in there.
(lathe, mill, cnc plasma, table hoist, eddy current retarded chassis dyno etc.) I would probably be better off on grid
but I am a bit of a hermit and power isn't available out there where I am building the house
Cheers,
Sam
-
sorry about all the questions. I am doing the strip down on some inspire 1.5 transformers at the moment
and I am considering doing a triple stack and doing either 4 or 5 layers on the secondary as usual I have a couple of questions.
what would a good place to start for number of turns on the primary and secondary be? (48v)
what size primary would you recommend/how many in hand?
how hard do you think I could safely push it?
thanks so much for all the help. without the help of this forum I would be screwed
Cheers,
Sam
-
Sam, before you rip it apart, find the 240v winding, and put 240v through it, wind 5 or 10 turns of any wire you have and measure the voltage, and divide by the turn number.. this gives you the turns/volt... from there you can work it all out.
I suspect about .9v/turn for the aerosharps, and 1.4 for inspire, but they use several transformers
.............oztules
-
it looks to be 1.2v I did 10 turns and got 12.16v from it. but the input was 250v would that make much of a difference?
also these are wound a bit different to the larger areosharp one I did. the aerosharp was wound 3 in hand all
at once but these inspires are would "oz style" lol 2 separate layers and then paralleled.
for some reason i can't post pics at the moment but i will later on.
Cheers,
Sam
-
There is the answer you were looking for 1.2v it is ( good enough )..... so 240v is 196 or close enough to 200 turns for the secondary, and your primary will be about 1/8 ( for 48v) or( 1/16 for 24v).. 200/8=25 turns for 48v unit, 12 turns for 24v unit.
..........oztules
-
would that be for a single core? if so when I use 3 of them do I just divide those numbers by 3?
thanks for the help.
Cheers,
Sam
-
i should be ready to start winding the triple stack transformer tonight. just a quick question. does 67 turns sound correct for the secondary?
Cheers,
Sam
-
yes, but I would do 70 to give some clearance from saturation... less idle current. ( thats for 3 cores @3.6v/turn )
...........oztules
-
beautiful. thanks for that. I glassed the cores together this morning and god damn it's a heavy mother now.
Cheers,
sam
-
the core are glassed together, a couple of layers of mylar down, the first layer of the secondary down a layer of resin and more mylar going to do another layer of the 2mm secondary today. man it is hell on your whole upper body, it doesn't help when the core alone weighs in at just on 35kg.
-
i have a fair few pics that i would like to add but it won't let me, i have tried with 2 browsers and i get the same resault. it uploads them and then says the page can't be contacted. is it just me? i have been trying for a few days
-
" man it is hell on your whole upper body"... aint that the truth!
Pics seem to work ok for me
Browse for file and then pick your "inline fullsize image" etc
Check your file type is site friendly.... ie .png .jpg etc.... and your size is in limits
.............oztules
-
some pics of the process so far.
-
got the pics sorted, i had turned up the quality on my camera and forgot. so the pictures were too big.
all sorted now.
Cheers,
Sam
-
a quick note on the recycled copper. i did some of it with the 3 bearing straightener and the rest with my hyper extreme ute straightening process ::).
and the stuff i pulled with the ute is waaayyy softer than the other stuff. just thought it may be of interest.
Cheers,
Sam
-
i did a test run on the inverter with the small transformer today, and it seemed to go ok. i havn't tried under load or anything as yet but i am happy with the idle current, it is 16.3w does that seem ok to you guys?. i still have a way to go but it is getting there. to start with i made a bit of a mistake and tried to power it from my two 24 volt battery chargers wired in series. that was a mistake, the voltage measured at 52v but when i switched it on it all seemed ok but then r2 (the one in the picture) started to smoke i looked at the meter quickly and shut it down, the meter was reading 72v :o. (wonder why it didn't shut down from over voltage?) i hooked it up to 4 old car batteries like i should have in the first place and i fired up fine. that resistor is running warm though is that normal?. the ecore is pretty noisy but i haven't glued it yet and the transformer is making a bit of noise but i am guessing that is because it isn't epoxied, i just stripped the primary off and added 20 turns to the seccondary?
thanks for all the help it sure is good when it all starts coming together ;D
Cheers,
Sam
-
one last question for now, haha
how many layers would you recommend on the secondary of this big trans i am making?
and what size primary?
i would like to make it as solid as practical
Cheers,
Sam
-
another thing is, I have read about bypassing current around the ct in many of your posts, but i can't find how you actually did it.
would you be able to shine some light on it for me please?
thanks,
Sam
-
just thought i should add that the wire i am using on the secondary is just over 1.8mm in diameter,
which according to the online calculators is 2.6mm squared
Cheers,
Sam
-
another layer down and epoxied. that's 3 hmm where to stop? ???
-
As you seem to want high power for long periods, three should do fine, but you will have room for 4 I suspect, so more never hurts... you only get to do it once.
How does that smaller single stack one perform now?
The pics are good, some text to explain them would help in their interpretation too.
The CT bypass is simply that... or neater and easier, is to change the resistor value on the output. You will see that under the board, and it generally looks like they have added it later on... they have... I think they make the same board and change the resistor for the expected wattage they want.... reduce that resistor, and it wlll damp the ct signal,... so more power.
Nice looking core there too... well done.
............oztules
-
thanks for the compliments, without your help none of this would have been possible.
i am hoping to do some better testing on the smaller one tomorrow when i bring some of my better batteries here from my block where we are building.
i will report on that tomorrow night.
do you have any suggestions on a good value to start with for the current shunt? or is it more of a trial and error thing,
would you think that 70mm2 would be adequate for the primary?
here are a few pics of the 15kw boards i have. i bought a complete 15kw inverter about a year ago and stripped it down for this project as the sellers at the time were refusing to ship me boards by themselves, but i figured compered to the price of other inverters that don't even seem to do half of what these can it was still cheap buying.
plus i now have the 3 transformers from it to play with. i am going to use one of them to experiment with the oil cooling idea. and am thinking of driving it with one of the control cards you are working on and one of the driver boards clockman and you are doing, could make for an interesting project.
anyway thanks again for sharing your knowledge with the world.
Cheers,
Sam
-
oops forgot the pics.
-
i just pulled the control board off, flipped it over and removed the tape, but i can't seem to find the resistor for the current bypass. what area of the board should i be looking in? i presumed it would be across the ct but i may be wrong.
Cheers,
Sam
-
Look to the front of the board for R109.... second in from the edge. If you think you have it, check for continuity from it to the CT output leads... then thats it.
70mmsq primary will be more than good
current shunt.... trial and error.. I suspect if you halve the resistance you have doubled the current needed to trigger.
................oztules
-
thanks for the quick reply. i will sus it out now.
Cheers,
Sam
-
that was the one. thanks for that.
do you think it would work if i removed that resistor, soldered 2 wires onto the pads then soldered them to a socket glued to the top of the board,
so that i can easily change the resistors without pulling too much apart?
Cheers,
Sam
-
do you think it would work if i removed that resistor, soldered 2 wires onto the pads then soldered them to a socket glued to the top of the board, so that i can easily change the resistors without pulling too much apart?
If you think about the way a CT works, you can deduce it will be fine, within reason.
A CT has a primary (usually a single turn), and a secondary - of some number of turns. Lets say it's 1000 turns.
For 10 amps through the primary, there will be 10/1000 amps in the secondary.
With no current, the voltage across the output will skyrocket (which is why you NEVER open-circuit the secondary of a CT!)
The burden resistor (load, or shunt, if you like) across the output of the CT determines the voltage across it, for a given current. So in the example above, 10A in a 1000:1 CT, means 10/1000 = 1/100A = 10mA
With a 1K resistor, you should see 10V. With a 100R resistor, you should see 1V
If the load resistor is very low value, then connection resistance will be an issue, but for any "sane" value of R, the connection resistance from your connector will be negligible.
-
... and if your lazy and can be bothered with the maths, a trim pot of 1000r will give you infinite control... but use cautiously.
I use the big welder, and trim down until I don't see the led flash...... and thats it.
...............oztules
-
thanks for the explanation rossw i had a rough idea of how the ct worked, but i did't know how the shunt fitted into the equation but it is clear to me now thanks to your excellent explanation. and thanks again oz for yet another great idea.
Cheers,
Sam
-
hi oz i saw on some of yours it look like pvc insulation tape instead of mylar. is it ok to use that instead?
-
It probably has the same melting point as your primary will.... but only as an outer wrap ( cosmetics), not for wrapping the secondaries... use the proper tape ....although if you don't have any.....better then nothing I guess.
Mine was cosmetic on the first ones I think.
...............oztules
-
hi guys,
i just did some testing on the small inverter and i'm not sure if all is well.
it powers normal loads ok. but if i try to start my compressor 3Hp belt drive it won't start output voltage drops to 109v and no alarm.
i fire up my bandsaw, oldschool .33hp belt drive alarm light flashes once and off she goes. so i decided to plug in my inverter ac dc tig welder so that i can vary the load with the foot controller, so i ramped it up slowly and the voltage holds stable at 238v right up until i am pulling 9.5a from the inverter then it drops instantly to 108v and the welder starts spluttering. i also tried my 2000w makita chop saw and on startup the voltage dips to 108v for a second then it plugs away fine.
from what i have read about this setup i thought it should have been a bit beefier than this, or i could just be wrong?
any ideas?
Cheers,
Sam
-
i swapped out the control card for the one on the 15kw board and it fired everything up fine and worked how i thought it should ;D and here comes the BUT.
but when i fired up my compressor for the second time i heard a crack and got that distinct burnt semi smell :( i can't see any burnt fet's but i am going through it now
it didn't go off with a big bang, no smoke but no output now
cheers,
Sam
-
there is no physical damage to the FET's but is there any way to do a rough check on them while they are still in the board? i have had a quick check and some of them are short between all 3 pins and others are open between all, not sure where to go from here?
thanks,
Sam
-
well all i can say is BANG!.
i would say that there must be something wrong with the control board, because i just strapped it onto my brand new 15kw mainboard and BANG.
any ideas where i should go from here, funds are a bit tight so i can't afford to just buy all new parts.
-
Good grief... the whole lot's is shot I'm afraid.
I suspect your smaller control card was intermittent, and that probably led to some fet problem in the fet baord... putting the 15kw controller may have pushed your probably by then fet board too far.... and then took out it's drivers.... I suspect the whole lot is toast now.
The control cards may be fixable if the damage did not go past the opto and totem pole driver transistors... in which case fixable if you able to do that sort of work.
The better news is the fet cards are easy to fix... simply replace all the fets or at least all the fets that are short... and they are usually only 70c/piece.
This is why I developed the last board.... in this case I replace three plug in chips, and the fets and we are fully renewed.... chips about $5 all up.... fets $24.00 for the set max..
If you try to fix the controller, then use 1 fet in each leg only, until you get it working properly, you don't want to kill 24 each time you mess up, and it is hard to test the control card out of the system
Normally you could drive those loads easily with the 15kw board and controller.... how small is your smaller unit?.. usually 5kw units will start anything anyway, and they are self protecting if the board is fully functional... your had a fault... you may get a new board/s for free if it is recently purchased.
I would urge you to try to build my little board, It seems to work fine, and you can tune your voltage and current points as you want..... I have not been able to kill a 8kw fet board with the new controller, no matter how hard I try to push it.
Remember, if you use the 002 design to remove the 393 and do the pin bypass on it, and change the resistors on the gates on the fet board to 5r6 and put diodes across them... read that whole thread a few times until you get the feel of what I did, what worked, and what did not.
Hopefully PJ will replace the boards free.
..............oztules
-
the smaller unit was a 6kw and is out of warranty by 10 days :( i have let the project drag out a bit long haha.
i am waiting to see if they will consider warranty on the 15kw unit. fingers crossed and i am seriously thinking that i might build a couple of your control cards seems like a much simpler design to troubleshoot should i have more problems in the future.
i have ordered new fets for the 15k unit and will work out what the resistor values are tomorrow and order them as well.
if they do the warranty on the 15k i will end up with a spare one so i won't bother fixing the 6k unit if it works out that way.
when i flicked the switch and saw the electrical forest fire i did think it was a pretty impressive show, then the anger kicked in hahaha
well now i'm off to read over your control board thread again, then start ordering
oh and i have some more pics that i will put up of my finished triple stack trans
Cheers,
Sam