Author Topic: Sine Wave Inverter  (Read 5325 times)

0 Members and 1 Guest are viewing this topic.

ibedonc

  • Full Member
  • ***
  • Posts: 113
Sine Wave Inverter
« on: September 27, 2004, 12:06:27 PM »
I am going to have a board house do my final version of the pcboards , I am making these so they plug into a card cage , they also can be run without a card cage


I can pay $245 for 2 boards or $453 for 20 , you will need 1 pwm control card and 1 output card , I am going to use 4 outputs  and have at least 2 spare


these will have 2 sides and solder mask , and silk screen


not trying to make anything off of them just trying to get the cost of my boards down

« Last Edit: September 27, 2004, 12:06:27 PM by (unknown) »

daleh007

  • Jr. Member
  • **
  • Posts: 62
Re: Sine Wave Inverter
« Reply #1 on: September 27, 2004, 01:19:19 PM »
Have you got a schematic and BOM (Bill of Materials)?

Daleh
« Last Edit: September 27, 2004, 01:19:19 PM by (unknown) »

John II

  • Full Member
  • ***
  • Posts: 119
Re: Sine Wave Inverter
« Reply #2 on: September 27, 2004, 02:03:43 PM »
Ibedonc, I'm highly interested ! Could you Email me at jjones2@inter-linc.net I need to discuss this further with you, but hate to clutter the board.


Thanks


John II


( In S.W. Missouri )

« Last Edit: September 27, 2004, 02:03:43 PM by (unknown) »

RatOmeter

  • Full Member
  • ***
  • Posts: 144
Re: Sine Wave Inverter
« Reply #3 on: September 27, 2004, 02:29:33 PM »
I would be interested, but can't promise to purchase at this time.  If you could swing the dollars and there's enough interest to justify it, the more you order, the lower the piece price.  Maybe you could get 50 for $650?  Also, you might shop around a little (if you haven't already).  A year or two ago I saw a PCB outfit out of Belarus or somesuch that made PCBs dirt-cheap.  Never used them though.


Also, I found out the hard way some time back what factors are the main cost drivers in PCB manufacture.  We had a backplane design that was 19"x5" and got prototypes (3 or 4 IIRC) that wound up costing about $55 each.  I called the PCB folks and asked them why they were so expensive and what I could do to bring the price down.  He said the odd size (19" long) required that they stop the manufacture in the middle of the process to rotate the material in the machine.  I redesigned it to shave 3" off and managed to cut the price by about 1/3.


The main cost drivers, most costly first, are (please correct any errors I may make here):



  1. any manual operations (like that repositioning thing I mentioned)
  2. Number of layers > 2
  3. number of holes/vias
  4. amount of material used.


Of course, custom requirements like copper thicker than one oz, special materials, complicated cutouts/shape, odd/non-standard tool sizes, very small features (IIRC 8 mils is the standard min feature size) will affect the end price.


You might talk to your PCB house and ask them what, if anything, you could do to lower the price.


Funny, the daughter boards that plugged into that backplane had to be redone.  The first batch were flawless per the design but for one little thing. The crappy old DOS-based Tango software I was using didn't have a pattern for a 68 pin PLCC socket in its library, so I very carefully, painstakingly created one... that was backward.  IOW, to make it work, I had to solder the socket on the 'wrong' side of the board.  Hard to hide that one; even a novice could see that it wasn't right.


Bummer.

« Last Edit: September 27, 2004, 02:29:33 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #4 on: September 27, 2004, 03:22:13 PM »
you would get both with the boards
« Last Edit: September 27, 2004, 03:22:13 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #5 on: September 27, 2004, 03:26:15 PM »
my cards are 8.5" x 4.625"


I am going to use pcbexpress.com, they have a max of 25 per run


the backplane I am going to make myself

only needs to be one sided and not very complicated

« Last Edit: September 27, 2004, 03:26:15 PM by (unknown) »

drdongle

  • Hero Member
  • *****
  • Posts: 552
Re: Sine Wave Inverter
« Reply #6 on: September 27, 2004, 04:27:39 PM »
I would be interisted in several PWM cards+ the programed micro controlers for same.


Carpe Vigor


Dr.D

« Last Edit: September 27, 2004, 04:27:39 PM by (unknown) »

deerslayer660

  • Newbie
  • *
  • Posts: 37
Re: Sine Wave Inverter
« Reply #7 on: September 27, 2004, 04:39:19 PM »
 

   I would really need to know total cost of all componants riscappotsicexc.

thats the only way to tell if project is worth time to put together dont like

spending 5 or 6 hrs putting something together to save 20$ or worse spending 20$

more please post when you have more info on price as it may only be a feasable

project if all componets are perchesed in bulk and sold as a kit I have seen things like a simple 7805 from nte sell for 3.50$  at radioshack 1.50 at the hamfest .50

 and so on but all in all you got my interest  george
« Last Edit: September 27, 2004, 04:39:19 PM by (unknown) »

deerslayer660

  • Newbie
  • *
  • Posts: 37
Re: Sine Wave Inverter
« Reply #8 on: September 27, 2004, 04:44:58 PM »
  sorry that schould read ris-cap-pots-ics-exc for some reason my slash disapears

when i post here it read good befor i clicked on post george
« Last Edit: September 27, 2004, 04:44:58 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #9 on: September 27, 2004, 04:47:38 PM »
this is not a micro based PWM , but both the sine ref and pwm are crystal based


the sine is generated with a ML2035 chip and the PWM is easy to find cmos and op-amp chips


the output uses ir2183 drivers and irfp360 mosfets


now I made this to run from a +- 170vdc , ether from batteries or a msw inverter rectified to + - 170vdc


I will be making a 12 - 48 vdc DC/DC later


this is a way to make your cheap msw inverter to a pure sine inverter


and the way I am doing it the DC/DC will be another card that will plug into the backplane


I may work on computers for a living but I do not like programing them :)

« Last Edit: September 27, 2004, 04:47:38 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #10 on: September 27, 2004, 05:04:37 PM »
the boards will be most of the cost , on the driver card the mosfets cost me $36 (12 x $3) the ir2183's are about $4  ( need 2) the rest is resistors /caps , need a 100 uh output inductor that will handle the amps, you need a 10uf oil filled A/C cap , got them for a few bucks off ebay


and a case to put it all in , I found a drive array chassis that I am making a new backplane for from surplus


but you do not have to put them in a card cage  also you could divide up your a/c circuits for how many cards you run


each one of my drivers in my setup should be able to do 100 amps  @ 240vac


but with 4 cards in case that would be about 50 amps per card cont.  of course these are max  and I have been doing about 1600 to 2200 kwh a month


my heat pump max is 22 kw  when the aux heat kicks in , but that mite only be a few days a year here in Texas

« Last Edit: September 27, 2004, 05:04:37 PM by (unknown) »

John II

  • Full Member
  • ***
  • Posts: 119
Re: Sine Wave Inverter
« Reply #11 on: September 27, 2004, 06:48:01 PM »
I would love to be in on the board purchases. Is the PWM control card going to cost you the same amount as the output board ?


I wouldn't mind owning a couple of PWM boards and four or more output boards. I would need to know the cost of what that would be.


A couple of questions though. Have you tested it against difficult to run inductive loads yet ? (just currious)


Also my largest concern.


I'm going to be needing to run this off of a 120vdc battery bank (100vdc to 160vdc) Do you have any idea how hard it would be to build a DC to DC converter to provide the + - 170vdc ? I think that Hilo90mhz who is on this forum was building a DC to DC converter to ? Any thoughts on this would really be appreciated.


John II

« Last Edit: September 27, 2004, 06:48:01 PM by (unknown) »

jasonweir

  • Newbie
  • *
  • Posts: 14
Re: Sine Wave Inverter
« Reply #12 on: September 27, 2004, 07:32:23 PM »
you mentioned a crystal based sin wave generation, how much work it would be to make it referenced off of grid power.  Then it could be used as a grid tie inverter.  With additional equipment of course


Thanks

Jason Weir

« Last Edit: September 27, 2004, 07:32:23 PM by (unknown) »

jasonweir

  • Newbie
  • *
  • Posts: 14
Re: Sine Wave Inverter
« Reply #13 on: September 27, 2004, 07:34:57 PM »
I am interested as well, will still need an approx total cost...


-J

« Last Edit: September 27, 2004, 07:34:57 PM by (unknown) »

LEXX

  • Full Member
  • ***
  • Posts: 129
Re: Sine Wave Inverter
« Reply #14 on: September 27, 2004, 08:44:22 PM »
I am very interested but I have one question: why did you design it to run off of +-170VDC?

LEXX
« Last Edit: September 27, 2004, 08:44:22 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #15 on: September 27, 2004, 09:32:53 PM »
you can already jumper it for ext. sync , not sure on best way to get line signal to 5 v pp for input
« Last Edit: September 27, 2004, 09:32:53 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #16 on: September 27, 2004, 09:45:30 PM »
actully you could run it on + - 130 to + - 180


but my reason is that I have a bunch of msw inverters and when you rectifiy the output like in would in a PC ATX power supply you get + - 169 volt ( I say 170 just rounding up )  so I am using the $200 2500 watt msw as a cheap dc/dc powersource also you can take 2 or more and add them together that way , but each msw needs it's own bridge


another reason is a am going to have a cap bank on the high side , I almost have 1 fard


so that when a motor starts it will get that from the caps , I also have 28 135 ah 12v batteries that I just got for Free , ( pulled early from our data center ) that I me run in series 13 x 13v ( float volts) = 169 volt x 2 13 for plus side , 13 for minus side


later I will design my own dc/dc's and sale the msw inverters  

« Last Edit: September 27, 2004, 09:45:30 PM by (unknown) »

TomW

  • Super Hero Member Plus
  • *******
  • Posts: 5130
  • Country: us
Re: Sine Wave Inverter
« Reply #17 on: September 27, 2004, 10:10:50 PM »
ibedonc;




not sure on best way to get line signal to 5 v pp for input


Just off the top of my head and without looking at the schematic, I would say a resistor or three and a couple of back to back zeners of proper size. Or even a string of standard diodes back to back to limit the voltage you sample. Of course this would give a chopped off wave not a sinewave. If that matters then a 24:1 transformer. They make 5 volt wall warts too for maybe a ready made solution. If its DC just bypass its diodes /caps etc. There is also a simple voltage divider network but that means the input voltage would need to be relatively stable to get 5 volts as needed.


Probably not what you need but just a couple ideas.


T

« Last Edit: September 27, 2004, 10:10:50 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #18 on: September 27, 2004, 10:25:16 PM »
here a picture of the surplus card cage (ex drive cage)




and this is the new hand made PWM board , man is tuff to do a double sided board with the laser printer method





so it you want to do it like mine then you will have to find something simlar , on ebay or computer surplus  , or if you have a metal shop available :)

« Last Edit: September 27, 2004, 10:25:16 PM by (unknown) »

linesrg

  • Newbie
  • *
  • Posts: 3
Re: Sine Wave Inverter
« Reply #19 on: September 28, 2004, 12:48:59 AM »
Ibedonc,

I've been following the development of your inverter as I'm very interested like some of the other respondents.

I'm definitely interested in at least the minimum necessary number of boards to make a one off device.

Have you considered using www.olimex.com in Bulgaria? They do some very good prices.

Regards
« Last Edit: September 28, 2004, 12:48:59 AM by (unknown) »

nothing to lose

  • Hero Member
  • *****
  • Posts: 1538
Re: Sine Wave Inverter
« Reply #20 on: September 28, 2004, 03:20:20 AM »
"( In S.W. Missouri )"


South Central here.

Ever go to Branson, 150 miles from me.

« Last Edit: September 28, 2004, 03:20:20 AM by (unknown) »

RatOmeter

  • Full Member
  • ***
  • Posts: 144
Re: Sine Wave Inverter
« Reply #21 on: September 28, 2004, 07:20:53 AM »
I think isolation is key here.  Like Tom said, a transformer could do the trick.  For syncing purposes, a zero crossing detector on the transformer output might be the easiest.


BTW, I'm really interesting in starting from your design and adding a microcontroller to it to control the PWM, line sync and other 'smarts'.  Can your schematic and PCB design file be made available? I'm itching to get back into embedded controls, at least a small project or two.

« Last Edit: September 28, 2004, 07:20:53 AM by (unknown) »

RatOmeter

  • Full Member
  • ***
  • Posts: 144
Re: Sine Wave Inverter
« Reply #22 on: September 28, 2004, 07:31:27 AM »
Thanks for that link!


Those are the folks I was talking about in an earlier post (Bularus/Bulgaria, I was kinda close).  I have mixed feelings about sending off a design to be made halfway around the world, but I've read good things about them.

« Last Edit: September 28, 2004, 07:31:27 AM by (unknown) »

Nando

  • Hero Member
  • *****
  • Posts: 1058
Re: Sine Wave Inverter
« Reply #23 on: September 29, 2004, 10:37:56 AM »
ibedonc:


More information, I would like to have, before I say yes, can you supply the basic schematic ?.


Can you define the protection circuitry ? For over current and for sudden drop of current at 80 % of full load capabilities.


What are the specifications of the Sine Wave Inverter ?.


Can you send the information directly ?>


Regards


Nando

nado37 at comcast dot net

« Last Edit: September 29, 2004, 10:37:56 AM by (unknown) »

Nando

  • Hero Member
  • *****
  • Posts: 1058
Re: Sine Wave Inverter
« Reply #24 on: September 29, 2004, 12:11:27 PM »
IBEDONC:


I would like to send to you an article that shows the schematic, BOM of a DC/AC converter for you to read a for you to examine your design in reference to possible errors in your design.


As a retired electronic engineer, power was one of my disciplines and just reading the little info yu have supplied I see problems which are CRITICAL that need to be tested carefully.

I still do consulting engineering in power design --I have done from low AC generation ( 5 watts ) to high power ( 500 KW +)


Please connect with me directly, I am not trying to put you down at all, I am trying to offer you assistance to solve unforeseen future problems that only experience may give the solutions to them, I congratulate for your effort -- long and hard, as well as, difficult --


Points to ponder IRP360 is a 400 volts 24 amps Mosfet-- you are using it for +- 170 volts supply.

Not good -- you need 500 volts minimum -- I could give you a complete dissertation why.


+
- 170 volts supply to generate 120 volts RMS or 170 Volts pick -- No overhead voltage for Mosfet saturation voltage and for choke current -- THINK why you need higher voltage, as a minimum you need +- 200 Volts -- the design should be for +- 220 Volts Peak DC, to allow for variations in the overall system, which forces you now to a new type of MosFets ( 500 Volts minimum).


There are many other things to ponder, like energy dumping at max power, temperature behavior, current protection for single AC source and for double AC source like yours, PCB and card cages, etc and more etc's.


Regards


Nando


nando37 at comcast dot net

« Last Edit: September 29, 2004, 12:11:27 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #25 on: September 29, 2004, 01:55:03 PM »
Hi , thanks for your comments , first , I will not release or sale (the pcboards) anything until I have working / tested unit , not even a schematic , high volts is nothing to play with


A lot of the points you pointed out I have thought about and will be in the final circuit  , and I know about the 170v pp for 120 but since I have not done the high volt tests yet , I was just thinking that 340 vdc would be enough to get 120v rms


in the next week , I will be setup to try the + - 170 supply


you can email me the info @ sinewave at star-c dot com


I have a lot of thought invested in this and over 20 years electronic repair , even in SMPS ( elpac,  etc ) that is why I do my stuff without CPU's , I have TAMU's design


that is why I was trying IGBT's first time ever using them .

« Last Edit: September 29, 2004, 01:55:03 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #26 on: September 29, 2004, 02:25:16 PM »
Hi , thanks for your comments , first , I will not release or sale (the pcboards) anything until I have working / tested unit , not even a schematic , high volts is nothing to play with


A lot of the points you pointed out I have thought about and will be in the final circuit  , and I know about the 170v pp for 120 but since I have not done the high volt tests yet , I was just thinking that 340 vdc would be enough to get 120v rms


in the next week , I will be setup to try the + - 170 supply


do you have a number I can call you ? can you email it to me ?


you can email me the info @ sinewave at star-c dot com


I have a lot of thought invested in this and over 20 years electronic repair , even in SMPS ( elpac,  etc ) that is why I do my stuff without CPU's , I have TAMU's design


that is why I was trying IGBT's first time ever using them .

« Last Edit: September 29, 2004, 02:25:16 PM by (unknown) »

ibedonc

  • Full Member
  • ***
  • Posts: 113
Re: Sine Wave Inverter
« Reply #27 on: September 29, 2004, 03:10:43 PM »
Hi , thanks for your comments , first , I will not release or sale (the pcboards) anything until I have working / tested unit , not even a schematic , high volts is nothing to play with


A lot of the points you pointed out I have thought about and will be in the final circuit  , and I know about the 170v pp for 120 but since I have not done the high volt tests yet , I was just thinking that 340 vdc would be enough to get 120v rms


in the next week , I will be setup to try the + - 170 supply


do you have a number I can call you ? can you email it to me ?


you can email me the info @ sinewave at star-c dot com


I have a lot of thought invested in this and over 20 years electronic repair , even in SMPS ( elpac,  etc ) that is why I do my stuff without CPU's , I have TAMU's design


that is why I was trying IGBT's first time ever using them .

« Last Edit: September 29, 2004, 03:10:43 PM by (unknown) »