Searching \ for '[EE] Zener as voltage dropper' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/power.htm?key=voltage
Search entire site for: 'Zener as voltage dropper'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Zener as voltage dropper'
2011\05\30@220628 by IVP

face picon face
part 1 1043 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

Hi all

Recycling again

re the attached

The max in-out differential voltage to the LM317 is 40V so I'd
want to drop a few volts off the supply, at least 5V

2N3055 gain is 20-70 @ 4A, transformer rated at 34V 6.5A

Vee will be a small -ve voltage so the output can go down to 0V

So, if I aimed for a Vout max of 38V (assuming regulator/transistor/
resistor losses will probably not quite reach that), at 6.5A that implies
a worst case base current of about 320mA, which will come from the
LM317/350, and will pass through the zener

Would a 6.8V 5W be a suitable dropper ? At less than full current
draw, it would need a load resistor to 'turn it on' ? Maybe 6.2V is
sailing a bit close to the wind

Would you include sharing resistors on the transistor bases ?

And then the diodes I've seen in some designs. What is the purpose
of the 3 x 1N4001 in place of the link to the bases. Simply a voltage
dropper for that particular design ? And the 1N5401 feedback from
Vout to supply ?

TIA

Joe

part 2 2326 bytes content-type:image/gif; name="v_control.gif" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2011\05\30@222412 by Adam Field

flavicon
face
> The max in-out differential voltage to the LM317 is 40V so I'd
> want to drop a few volts off the supply, at least 5V
>
> 2N3055 gain is 20-70 @ 4A, transformer rated at 34V 6.5A
>
> Vee will be a small -ve voltage so the output can go down to 0V
>
> So, if I aimed for a Vout max of 38V (assuming regulator/transistor/
> resistor losses will probably not quite reach that), at 6.5A that implies
> a worst case base current of about 320mA, which will come from the
> LM317/350, and will pass through the zener

The LM317 doesn't need to be ground referenced, and there are 60V
versions available from mouser (512-LM317AHVT).

You could also use two LM317s in a "tracking pre-regulator" found in
the app notes of the LM317 from TI:

http://focus.ti.com/lit/ds/symlink/lm317.pdf

Information on floating the LM317 is available, in this mode you can
regulate from Vin to (Vin-LM317) :

http://www.national.com/ms/LB/LB-47.pdf

I would try to combine any of these with a pass transistor rather than
using a zener to protect the LM317

2011\05\30@222943 by RussellMc

face picon face
> . What is the purpose
> of the 3 x 1N4001 in place of the link to the bases.

Reduce output x 3 x Vd ~= 1.8V.
Vout can be V_LM317 - 3 x Vd - Vbe_2N3055 ~~~~= 0V min if Vee is grounded..

If Vee is < ground then cct can go to ground without the diodes  (maybe).


ie As LM317 goes to minimum of 1.25V this allows Vout to go to ~= true Zero.

>  And the 1N5401 feedback from
> Vout to supply ?

That's a reverse discharge diode for when Vin is removed and you may
have larghe caps on the output (not shown).
Without this diode, if Vin is interrupted the cct may be reverse
biased by whatever Vout was - which may or may not cause damage
otherwise.

__

Under low to no load with 34VAC then Vincap can go to 34 x sqrt(2)~= 48V.
R plus zener will clamp this.
Deep-ending on transformer etc power loss in series R may become significant.
At 320 mA losses are ~= 100 mW per ohm of dropper or 0.32 W per volt of dropper.


2011\05\30@232122 by IVP

face picon face
> I would try to combine any of these with a pass transistor rather
> than using a zener to protect the LM317.

Thanks, will look those links u

2011\05\30@232124 by IVP

face picon face
>> of the 3 x 1N4001 in place of the link to the bases.
>
> Reduce output x 3 x Vd ~= 1.8V.
> Vout can be V_LM317 - 3 x Vd - Vbe_2N3055 ~~~~= 0V min
> if Vee is grounded.
>
> If Vee is < ground then cct can go to ground without the diodes
> (maybe).

Ah. I was looking at the wrong end. It wasn't to reduce Vout, it was to
reduce base Vmin ie compensate for the 1.25V min of the LM317. I
think I'd prefer a Vee, as changing base current will change the Vdrop
of the diodes

>>  And the 1N5401 feedback from
>> Vout to supply ?
>
> That's a reverse discharge diode for when Vin is removed and you
> may have larghe caps on the output (not shown)

Similar to reverse diode across LM78xx. It would be wise then to
include it. Although the PSU itself doesn't have any large filter caps,
the target circuit might have

> Under low to no load with 34VAC then Vincap can go to 34 x
> sqrt(2)~= 48V. R plus zener will clamp this

So you're suggesting the more common arrangement of a series R
to ZD k, ZD a to ground. eg a 24 ohm 5W + 39V 5W zener

46.5 - 39 = 7.5, I = 0.32A

V = I * R, R = 7.5 / 0.32 = 23.4R
P = 0.32 * 0.32 * 24 = 2.46

2011\05\31@002911 by RussellMc

face picon face
> >> of the 3 x 1N4001 in place of the link to the bases.
> >
> > Reduce output x 3 x Vd ~= 1.8V.
> > Vout can be V_LM317 - 3 x Vd - Vbe_2N3055 ~~~~= 0V min
> > if Vee is grounded.
> >
> > If Vee is < ground then cct can go to ground without the diodes
> > (maybe).

> Ah. I was looking at the wrong end. It wasn't to reduce Vout, it was to
> reduce base Vmin ie compensate for the 1.25V min of the LM317. I
> think I'd prefer a Vee, as changing base current will change the Vdrop
> of the diodes

Add an opamp and feedback (as per other similar ccts recently) and you
get a stable result.
Diodes do dropping to true zero in driver and opamp allows reference
to compare true output to reference.


          R


2011\05\31@014927 by IVP

face picon face
> Add an opamp and feedback (as per other similar ccts recently) and
> you get a stable result.
> Diodes do dropping to true zero in driver and opamp allows reference
> to compare true output to reference.

OK, noted

Option B is that there'll be a PIC driving at least one 3-digit LED
display. It would be little bother to make Vee with a spare pi

2011\05\31@112505 by Denny Esterline

picon face
{Quote hidden}

Recycling is admirable, but I think you're trying to drive a screw with a
hammer.
At this power level I'd strongly recommend some type of switching regulator,
or even a switching pre-regulator if you have compelling reasons to use
linear. There's lots of good references about, look at "Roman Black's"
regulator circuit if you want a small discrete component design, or the
National Semiconductor "Simple Switcher" line if you'd rather have a canned
solution.

-Denn

2011\05\31@191355 by IVP

face picon face

> Recycling is admirable, but I think you're trying to drive a screw with a
> hammer

Yup, ordinarily I'd agree, but I've accumulated so many parts over the
years that occassionally you just need to put some together otherwise
it wasn't worth hanging on to them. I do tend to not accumulate anymore
unless it's something a bit special. On Friday I was visiting a workshop
and noticed in their bin part of a marine cockpit panel. That was a grab
because it had a couple of $100 worth of immediately useful - given the
current projects - power switches and fuse holders

> At this power level I'd strongly recommend some type of switching
> regulator, or even a switching pre-regulator if you have compelling
> reasons to use linear

One reason is that they're there, unused

I do have quite a variety of switching components and they get used
from time to time, and of course I still buy new components. I've not
finished yet with PSUs and have taken your comments on board

Jo


'[EE] Zener as voltage dropper'
2011\06\03@205945 by IVP
face picon face
part 1 258 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

I finished one of the PSUs, this one a 0-38V 12A

3 x 1N4001 does reduce the LM317's base drive to 0V, so
Vee was not needed. A "6k72" pot gets Vout to maximum
with no ineffective top end of the track

part 2 2621 bytes content-type:image/gif; name="v_control2.gif" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2011\06\03@210924 by IVP

face picon face
part 1 600 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

I finished one of the PSUs, this one a 0-38V 12A. All parts were
in the junk box, so it's a satisfying freebie. And this is one of those
times when it *is* an advantage to be able make your own PCBs

Wired to Vout is a PIC-based 4-digit LED display, also from parts
in the junk box. I used 7:1 input divider, 5.12V Vref and s/w to make
an FSD of 40.96V (40mV resolution) from the ADC result

3 x 1N4001 does reduce the LM317's base drive to 0V, so Vee
was not needed. A "6k72" pot gets Vout to maximum with no
ineffective top end of the track

part 2 2621 bytes content-type:image/gif; name="v_control2.gif" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2011\06\05@214522 by Jonathan Hallameyer

picon face
> Recycling is admirable, but I think you're trying to drive a screw with a
> hammer.
> At this power level I'd strongly recommend some type of switching regulator,
> or even a switching pre-regulator if you have compelling reasons to use
> linear. There's lots of good references about, look at "Roman Black's"
> regulator circuit if you want a small discrete component design, or the
> National Semiconductor "Simple Switcher" line if you'd rather have a canned
> solution.
>
> -Denny
>

More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...