Re: 9VAC from 5VDC?

From: john/lori <lgnjh_at_earthlink.net>
Date: Fri, 13 Sep 2013 18:25:40 -0700
Message-ID: <5233BB14.1010904@earthlink.net>
Jim Brain wrote:
> On 9/13/2013 2:57 PM, Gerrit Heitsch wrote:
> 
>> On 09/13/2013 09:15 PM, silverdr@wfmh.org.pl wrote:
>>
>>>
>>> On 2013-09-13, at 05:33, Jim Brain wrote:
>>>
>>>> Does anyone have ideas (or could point to a person or company) that 
>>>> might have a way to safely synthesize an isolated 9VAC for the 
>>>> machines from 5VDC?
>>>
>>>
>>> Yes, that's one of the things, I'd like to see solved for good as 
>>> well. I was approaching it from different angles. Even "invented" PWM 
>>> based sine-wave generation once before sleep ;-) The problem always 
>>> was that in the end one needs a transformer, which is what I always 
>>> wanted to avoid in the first place..
>>
>>
>> You can't. There is no way around a transformer in this case.
> 
> 
> I'm not opposed to a transformer.  I have, I think, the other piece
> 
> o uC DSS implementation (2 of them, one is a R2R implementation and 1 is 
> a PWM implementation), coupled with a power transistor and a RC filter 
> to smooth out the waveform.
> 
> Now, if I could just find a 1:2 transformer that would handle 9VAC (9W?) 
> and work at 50 and 60 Hz, I'd be golden.
> 
> But, I know *NOTHING* about inductors and transformers.  I faintly 
> remember taking a part of a course on them in college, but nothing stuck.
> 
> So, Bil et al, help would be appreciated.  I looked at Coilcraft, and 
> saw some step down and forward mode transformers, but I need the 
> reverse, I think.
> 
> Jim
> 
> 
>       Message was sent through the cbm-hackers mailing list
> 

Magnetics are traditionally black magic, but transformers aren'
all that hard.

I assume you're talking 5VDC so you want a transformer that's
more like 2.6 : 1 (9VAC is 12.73 peak)

Assuming you're dead set against just using a 9VAC wallwart or
something next to your 5VDC, I think I'd try 5VDC to ~ +-13VDC
then synthesize the 9VAC from that.

The cheap inverters use a wave form something like this.

    _       _       _
  _| |_   _| |_   _| |_
|     |_|     |_|

Typically (I think) with the area under the curve aproximating
a sine wave and the peak close to the sine wave peak.

If you were going to wind your own transformer you could give it
more steps.

If you were using a computer supply with +-12 and +-5 VDC
you could try synthesizing it from that (maybe filtering)
then running it thru a 1:1 isolation transformer.
It would depend on how much power you want, PC supplies
usually aren't rated for much current on the negative supplies.



       Message was sent through the cbm-hackers mailing list
Received on 2013-09-14 02:00:05

Archive generated by hypermail 2.2.0.