Searching \ for '[EE] poor mans negative rail generation' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=poor+mans+negative
Search entire site for: 'poor mans negative rail generation'.

Exact match. Not showing close matches.
PICList Thread
'[EE] poor mans negative rail generation'
2010\02\26@104709 by alan smith

picon face
OK, so I have a part that requires a +/-12V on it, and I do have +24V available.  So, a poor mans method would be to use a voltage divider and generate a +/- 12V and use the centertap for the ground reference.  The 'ground' or virtual ground, would be completely isolated from the rest of the circuit of course.  The kicker is that there are digital signals (SPI), referenced back to the board ground, and the part requires a digital 3.3V for this interface (actually can be 5V).  I could generate a local 5V from the +12, referenced to the virtual ground again but the only way I can see to get the SPI is to optically/transformer couple it into this section.  I believe ADI makes a part that can do this, and its not the fact that its another part, its also room on the board for another part.

Short of this idea, what other methods to generate a split supply from the +24V ?  I have very little room for much of anything, so it has to be small and pretty sure not enough room for a negative buck switcher unless very tiny.  Total current isnt going to exceed 1A


     

2010\02\26@111846 by Spehro Pefhany

picon face
At 10:47 AM 26/02/2010, you wrote:
>OK, so I have a part that requires a +/-12V on it, and I do have
>+24V available.  So, a poor mans method would be to use a voltage
>divider and generate a +/- 12V and use the centertap for the ground
>reference.  The 'ground' or virtual ground, would be completely
>isolated from the rest of the circuit of course.  The kicker is that
>there are digital signals (SPI), referenced back to the board
>ground, and the part requires a digital 3.3V for this interface
>(actually can be 5V).  I could generate a local 5V from the +12,
>referenced to the virtual ground again but the only way I can see to
>get the SPI is to optically/transformer couple it into this
>section.  I believe ADI makes a part that can do this, and its not
>the fact that its another part, its also room on the board for another part.
>
>Short of this idea, what other methods to generate a split supply
>from the +24V ?  I have very little room for much of anything, so it
>has to be small and pretty sure not enough room for a negative buck
>switcher unless very tiny.  Total current isnt going to exceed 1A

What do you need the +/-V for? Could you use a "bridging amplifier"?
(ie. the load would not be ground referenced, but would see up to +/- 24
less whatever the amplifiers (one @ near 0V and one @ near 24V-- those
values added together) need to avoid saturation).

Downside is that you need two amplifiers (one for each side of the load.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
spam_OUTspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com



2010\02\26@112104 by Harold Hallikainen

face
flavicon
face
I wonder if you could capacitively couple the SPI signals. CSn spends most
of its time high, so put a pull-up on the SPI chip. Set up your SPI so
clock and data spend most of their time low, then put pull-downs at the
SPI chip. Put a capacitor and a series resistor (to limit current at power
up) on each line between the PIC and the "high" SPI chip. The capacitors
will charge to the fixed offset voltage through the clamp diodes on power
up.

Might work...

Harold



--
FCC Rules Updated Daily at http://www.hallikainen.com - Advertising
opportunities available!

2010\02\26@112134 by M.L.

flavicon
face
On Fri, Feb 26, 2010 at 11:47 AM, alan smith <.....micro_eng2KILLspamspam@spam@yahoo.com> wrote:

> OK, so I have a part that requires a +/-12V on it, and I do have +24V
> available.  So, a poor mans method would be to use a voltage divider and
> generate a +/- 12V and use the centertap for the ground reference.  The
> 'ground' or virtual ground, would be completely isolated from the rest of
> the circuit of course.  The kicker is that there are digital signals (SPI),
> referenced back to the board ground, and the part requires a digital 3.3V
> for this interface (actually can be 5V).  I could generate a local 5V from
> the +12, referenced to the virtual ground again but the only way I can see
> to get the SPI is to optically/transformer couple it into this section.  I
> believe ADI makes a part that can do this, and its not the fact that its
> another part, its also room on the board for another part.
>
> Short of this idea, what other methods to generate a split supply from the
> +24V ?  I have very little room for much of anything, so it has to be small
> and pretty sure not enough room for a negative buck switcher unless very
> tiny.  Total current isnt going to exceed 1A
>
>
If you need an amp from your negative supply then splitting the rails with
an analog divider (BJT emitter follower fed by virtual ground referenced
OPAMP) will dissipate too much power to be useful. The most reasonable
solution is a buck-boost supply. This will also give you a true ground so
you don't have to jump through hoops to get your digital signals in and out.

--
Martin K.

2010\02\26@113415 by alan smith

picon face
The reason is this chip is used to generate a 0-10V or 4-20mA drive, in the same part (saving space for doing the same thing with two designs).
Running on a single rail generates 'dead bands' according to what I got back.

--- On Fri, 2/26/10, Spehro Pefhany <speffspamKILLspaminterlog.com> wrote:

{Quote hidden}

> -

2010\02\26@131731 by Michael Rigby-Jones

flavicon
face


> -----Original Message-----
> From: @spam@piclist-bouncesKILLspamspammit.edu [KILLspampiclist-bouncesKILLspamspammit.edu] On
Behalf
> Of alan smith
> Sent: 26 February 2010 15:47
> To: Microcontroller discussion list - Public.
> Subject: [EE] poor mans negative rail generation
>
> OK, so I have a part that requires a +/-12V on it, and I do have +24V
> available.  So, a poor mans method would be to use a voltage divider
and
> generate a +/- 12V and use the centertap for the ground reference.
The
> 'ground' or virtual ground, would be completely isolated from the rest
of
> the circuit of course.  The kicker is that there are digital signals
(SPI),
> referenced back to the board ground, and the part requires a digital
3.3V
> for this interface (actually can be 5V).  I could generate a local 5V
from
> the +12, referenced to the virtual ground again but the only way I can
see
> to get the SPI is to optically/transformer couple it into this
section.  I
> believe ADI makes a part that can do this, and its not the fact that
its
> another part, its also room on the board for another part.
>
> Short of this idea, what other methods to generate a split supply from
the
> +24V ?  I have very little room for much of anything, so it has to be
> small and pretty sure not enough room for a negative buck switcher
unless
> very tiny.  Total current isnt going to exceed 1A

You need to make absolutely sure that there will be no significant
digital currents introduced into your virtual ground as any SPI noise on
this will come straight out of the device as a voltage or current.  

You may be able to use current mirrors to do the voltage translation;
they are often very handy for this kind of thing IME.

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2010\02\26@133931 by alan smith

picon face
Well, sounds like to me, since this is an industrial product, that the best way to do this is a negative buck regulator.  Just have to find room for it someplace.

Thanks for all the replies

--- On Fri, 2/26/10, Michael Rigby-Jones <RemoveMEMichael.Rigby-JonesTakeThisOuTspamoclaro.com> wrote:

> From: Michael Rigby-Jones <spamBeGoneMichael.Rigby-JonesspamBeGonespamoclaro.com>
> Subject: RE: [EE] poor mans negative rail generation
> To: "Microcontroller discussion list - Public." <TakeThisOuTpiclistEraseMEspamspam_OUTmit.edu>
> Date: Friday, February 26, 2010, 10:17 AM
>
>
> > {Original Message removed}

2010\02\26@191312 by Xiaofan Chen

face picon face
On Sat, Feb 27, 2010 at 12:34 AM, alan smith <RemoveMEmicro_eng2spamTakeThisOuTyahoo.com> wrote:
> The reason is this chip is used to generate a 0-10V or 4-20mA drive,
>  in the same part (saving space for doing the same thing with two designs).
> Running on a single rail generates 'dead bands' according to what I got back.

Yes that is when your accuracy requirement is on the high side.
Use a rail to rail input amplifier may help. 4...20mA should be
okay. 0...10V is the issue. 0...20mA (if you need that) can be
an issue as well.

Often we need to provide -10V...10V along with 0...20mA
and 4...20mA, in that case, we need to have a +/-15V anyway.
But when there is only 0...20mA and 4...20mA requirement,
sometimes you can get rid of the negative supply if the
requirement is not too high. When the requirement is
high, then we have to have a negative -3V (or more
negative) supply so you really get 0mA when you mean
0mA.

By the way, now ADI (and Maxim) has nice output driver
which can product 0...10V and 4...20mA (and other
ranges). They are nice product.

As for the SPI digital isolation, the digital isolator from
ADI and TI are very good. They are more expensive
than the high speed opto coupler but consume much
less current.



--
Xiaofan http://mcuee.blogspot.com

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...