230VAC/50Hz -> 115VAC/60Hz converter

[Squidster]

Hello,
I'm trying to design a 230VAC/50Hz to 115VAC/60Hz converter. This is
specifically for applications that are frequency sensitive.

Specs:
Input: 220-240VAC/50Hz (plusminus 1Hz) @10A
Output1: 110-120VAC/50Hz (plusminus 1Hz) @20A \Combined is 20A
Output2: 110-120VAC/60Hz (plusminus 1Hz) @20A /

Here's the blocks:
1. 230VAC/50Hz input, offline rectified and filtered to 340VDC.
2. Chopped up (20-50Khz) either via a BJT or MOSFET power switch with PWM
controller (either 555 or PWM IC).
3. PWM wave fed into a high-frequency transformer (2:1)
4. Rectified secondary with schottky power diodes to recover DC.
5. This then is pass throught the 2nd. stage low-frequency choppers one
for 50Hz and one for 60Hz.
6. After going through the 50Hz and 60Hz choppers; we pass through a final
stage of low-pass-filter (LC, with fT=50 and fT=60 respectively)
7. Feedback from output fedback into the PWM controller.

Is there anything fundamentally wrong with the design?
I went through 2 stages because I wanted to avoid a huge mains transformer
rated at 20A!

Comments, caveats, experiences, component choices, are welcome!
I need some further design help on the:
- high-frequency 2:1 transformer
- PWM controller (IC recommendations?)
- chopper drive

Thanks!
/wai-sun

 

[[email protected]]

Hello,
I'm trying to design a 230VAC/50Hz to 115VAC/60Hz converter. This is
specifically for applications that are frequency sensitive.

Specs:
Input: 220-240VAC/50Hz (plusminus 1Hz) @10A
Output1: 110-120VAC/50Hz (plusminus 1Hz) @20A \Combined is 20A
Output2: 110-120VAC/60Hz (plusminus 1Hz) @20A /

Must output be galvanicly isolated or something?

Here's the blocks:
1. 230VAC/50Hz input, offline rectified and filtered to 340VDC.

filter = capacitors?

Extra capacitor network with a semiconductor in parallel with mains to
perform active powerfactor correction?
(needs feedback too

2. Chopped up (20-50Khz) either via a BJT or MOSFET power switch with PWM
controller (either 555 or PWM IC).
3. PWM wave fed into a high-frequency transformer (2:1)

4. Rectified secondary with schottky power diodes to recover DC.
5. This then is pass throught the 2nd. stage low-frequency choppers one
for 50Hz and one for 60Hz.
6. After going through the 50Hz and 60Hz choppers; we pass through a final
stage of low-pass-filter (LC, with fT=50 and fT=60 respectively)
7. Feedback from output fedback into the PWM controller.

Is there anything fundamentally wrong with the design?
I went through 2 stages because I wanted to avoid a huge mains transformer
rated at 20A!

Comments, caveats, experiences, component choices, are welcome!
I need some further design help on the:
- high-frequency 2:1 transformer pulvercore?

- PWM controller (IC recommendations?)
- chopper drive

Maybe IBGT will do the job?
You could even do sinewave pwm and have sine output..

 

[ChrisGibboGibson]

Hello,
I'm trying to design a 230VAC/50Hz to 115VAC/60Hz converter. This is
specifically for applications that are frequency sensitive.

Specs:
Input: 220-240VAC/50Hz (plusminus 1Hz) @10A
Output1: 110-120VAC/50Hz (plusminus 1Hz) @20A \Combined is 20A
Output2: 110-120VAC/60Hz (plusminus 1Hz) @20A /

Here's the blocks:
1. 230VAC/50Hz input, offline rectified and filtered to 340VDC.
2. Chopped up (20-50Khz) either via a BJT or MOSFET power switch with PWM
controller (either 555 or PWM IC).
3. PWM wave fed into a high-frequency transformer (2:1)
4. Rectified secondary with schottky power diodes to recover DC.
5. This then is pass throught the 2nd. stage low-frequency choppers one
for 50Hz and one for 60Hz.
6. After going through the 50Hz and 60Hz choppers; we pass through a final
stage of low-pass-filter (LC, with fT=50 and fT=60 respectively)
7. Feedback from output fedback into the PWM controller.

Is there anything fundamentally wrong with the design?
I went through 2 stages because I wanted to avoid a huge mains transformer
rated at 20A!

Comments, caveats, experiences, component choices, are welcome!
I need some further design help on the:
- high-frequency 2:1 transformer
- PWM controller (IC recommendations?)
- chopper drive

What you are proposing is perfectly feasable however...

One or two of your statements above imply that you're not very experienced with
this type of equipment.

Even for someone with a lot of experience on very high power switch mode
equipment I can see 12 to 18 months development work in this project.

You still want to get involved ?

Gibbo

 

[Rich Grise]

Hello,
I'm trying to design a 230VAC/50Hz to 115VAC/60Hz converter. This is
specifically for applications that are frequency sensitive.

Specs:
Input: 220-240VAC/50Hz (plusminus 1Hz) @10A
Output1: 110-120VAC/50Hz (plusminus 1Hz) @20A \Combined is 20A
Output2: 110-120VAC/60Hz (plusminus 1Hz) @20A /

Here's the blocks:
1. 230VAC/50Hz input, offline rectified and filtered to 340VDC.
2. Chopped up (20-50Khz) either via a BJT or MOSFET power switch with PWM
controller (either 555 or PWM IC).
3. PWM wave fed into a high-frequency transformer (2:1)
4. Rectified secondary with schottky power diodes to recover DC.
5. This then is pass throught the 2nd. stage low-frequency choppers one
for 50Hz and one for 60Hz.
6. After going through the 50Hz and 60Hz choppers; we pass through a final
stage of low-pass-filter (LC, with fT=50 and fT=60 respectively)
7. Feedback from output fedback into the PWM controller.

Is there anything fundamentally wrong with the design?

Yes.

1. It's obviously homework. We don't do other people's homework for
free.

2. This is a terribly stupid approach to this problem, when you can
get a 50 HZ synchronous motor and couple it to a 115V 60 Hz generator,
for about one-tenth the time and money.

Have Fun!
Rich

 

[Frank Bemelman]

Hello,
I'm trying to design a 230VAC/50Hz to 115VAC/60Hz converter. This is
specifically for applications that are frequency sensitive.

Specs:
Input: 220-240VAC/50Hz (plusminus 1Hz) @10A
Output1: 110-120VAC/50Hz (plusminus 1Hz) @20A \Combined is 20A
Output2: 110-120VAC/60Hz (plusminus 1Hz) @20A /
[snip]

Is there anything fundamentally wrong with the design?
I went through 2 stages because I wanted to avoid a huge mains transformer
rated at 20A!

I don't know, but there is at least something fundamentally wrong
with the specs. 100% efficiency is impossible. Two stages at 80%
gives 65%.

 

[Tim Wescott]

Yes.

1. It's obviously homework. We don't do other people's homework for
free.

2. This is a terribly stupid approach to this problem, when you can
get a 50 HZ synchronous motor and couple it to a 115V 60 Hz generator,
for about one-tenth the time and money.

Have Fun!
Rich

1. If it were homework it'd be better framed (see Mr. Bemelman's
comment about 100% efficiency).

2. I suspect that a solid-state solution would be smaller and cheaper
_if_ you're doing this for production -- but then I suspect that
_someone_ makes these already. The synchronous motor/60Hz generator
idea may work quite well if you can get something surplus.

If I were going to do this I'd also investigate just generating the
120VAC straight off of the 340VDC rail. As pointed out by Gibbo, you're
going to do a _lot_ of work on this -- it'd be a good vehicle for
learning power switching electronics (if you don't kill yourself), but
you should get help if you want to do this for production, or look for
an existing converter if you just need it.

 

[Pooh Bear]

4. Rectified secondary with schottky power diodes to recover DC.


Comments, caveats, experiences, component choices, are welcome!

Specifically, Shottky diodes are normally only suitable for low voltage.

Those filters are going to be damn large too. Might be better to generate the
50 & 60 Hz using ( say 30kHz ) PWM rather than chopping.


Graham

 

[Tim Wescott]

Pooh Bear wrote:

--snip--

Specifically, Shottky diodes are normally only suitable for low voltage. -- snip--
Graham

http://www.cree.com/ftp/pub/CSD20060.pdf

 

[Spehro Pefhany]

Pooh Bear wrote:

--snip--


http://www.cree.com/ftp/pub/CSD20060.pdf

Hmm... I wonder if they think it's a liability to mention that it's a
Silicon Carbide device?


Best regards,
Spehro Pefhany

 

[Tim Wescott]

Hmm... I wonder if they think it's a liability to mention that it's a
Silicon Carbide device?


Best regards,
Spehro Pefhany

Good question -- I don't do any design at 600V, but I question how much
better a SiC shottkey with a 1.5V forward drop is over a silicon diode.
I suppose the recovery time can't be beat...

 

[Mike]

Who's to say your solution is perfect and not "stupid" also? Before drawing
a conclusion like that, I think you should have evaluated the requirements
(e.g. perhaps he wanted a portable converter; not a clumsy mechanical one).

Mike

 

[Pooh Bear]

Pooh Bear wrote:

--snip--


http://www.cree.com/ftp/pub/CSD20060.pdf

Sadly the forward voltage makes it rather less helpful in this application
than 'normal' schottkies.


Graham

 

[Tim Wescott]

Tim Wescott wrote:




Sadly the forward voltage makes it rather less helpful in this application
than 'normal' schottkies.


Graham

Well, I view that page as kind of a party trick -- something to pull out
of my pocket whenever someone mentions that schottkeys are low voltage.

But it's not a bad forward voltage for something that'll hold off 600V,
and it should have really good reverse recovery characteristics, not to
mention thermal properties.

 

[N. Thornton]

2. This is a terribly stupid approach to this problem, when you can
get a 50 HZ synchronous motor and couple it to a 115V 60 Hz generator,
for about one-tenth the time and money.

True, but there are issues with rotary convertors.
Lack of voltage stability
Lack of frequency stability
weight
noise
non-instant start and stop
inductive source
and efficiency typ. in the region of 65%

It could be just the thing in some cases of course, it all depends on the app.

More app info would help, as there may be simpler app-specific options too.


NT