Parallel Voltage Regulators....
Archiver
Posts: 46,084
Thanks for the responses guys. Richard, what you mentioned here is what was
in the back of my mind bugging me - can any of you expound on this a bit more?
Tim
Timothy Medema
CrystaLite, Inc.
3307 Cedar St. (425) 745-6000 800-666-6065
Everett, WA 98201 Fax: (425) 257-0232
www.crystaliteinc.com
<mailto:timm@c...>timm@c...
The information transmitted is intended only for the person or entity to
whom it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you received
this in error, please contact the sender and delete the material from any
computer.
[noparse][[/noparse]Non-text portions of this message have been removed]
in the back of my mind bugging me - can any of you expound on this a bit more?
Tim
Timothy Medema
CrystaLite, Inc.
3307 Cedar St. (425) 745-6000 800-666-6065
Everett, WA 98201 Fax: (425) 257-0232
www.crystaliteinc.com
<mailto:timm@c...>timm@c...
The information transmitted is intended only for the person or entity to
whom it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you received
this in error, please contact the sender and delete the material from any
computer.
[noparse][[/noparse]Non-text portions of this message have been removed]
Comments
it might have been for pass transistors on a power supply. There were
several 2n3055 in parallel to get 20-30 amps on a 12 volt power supply.
Since the transistors are not manufactured 100% exactly the same there are
slight differences. If the internal resistances are slightly different then
the good transistor carries more than its share of the load and the loafing
transistor doesnt do its fair share.
The idea was to put a very low resistor in each output terminal connection
before tying them all together. It seems that they were 0.33 ohms and rated
for about 10 watts. The extra resistance added a small voltage drop but it
was worth it to get all transistors to carry their fair share of the load.
maybe this applies to regulators also ?
richard
Original Message
From: "Timothy Medema" <timm@c...>
To: <basicstamps@yahoogroups.com>
Sent: Monday, October 01, 2001 11:11 AM
Subject: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
> Thanks for the responses guys. Richard, what you mentioned here is what
was
> in the back of my mind bugging me - can any of you expound on this a bit
more?
>
> Tim
>
>
>
> Timothy Medema
> CrystaLite, Inc.
> 3307 Cedar St. (425) 745-6000 800-666-6065
> Everett, WA 98201 Fax: (425) 257-0232
>
> www.crystaliteinc.com
> <mailto:timm@c...>timm@c...
>
>
> The information transmitted is intended only for the person or entity to
> whom it is addressed and may contain confidential and/or privileged
> material. Any review, retransmission, dissemination or other use of, or
> taking of any action in reliance upon, this information by persons or
> entities other than the intended recipient is prohibited. If you received
> this in error, please contact the sender and delete the material from any
> computer.
>
> [noparse][[/noparse]Non-text portions of this message have been removed]
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>it might have been for pass transistors on a power supply. There were
>several 2n3055 in parallel to get 20-30 amps on a 12 volt power supply.
>Since the transistors are not manufactured 100% exactly the same there are
>slight differences. If the internal resistances are slightly different then
>the good transistor carries more than its share of the load and the loafing
>transistor doesnt do its fair share.
>
>The idea was to put a very low resistor in each output terminal connection
>before tying them all together. It seems that they were 0.33 ohms and rated
>for about 10 watts. The extra resistance added a small voltage drop but it
>was worth it to get all transistors to carry their fair share of the load.
>
>maybe this applies to regulators also ?
That is true of regulators too. The output voltage of one will be a
little higher than another. The one that has the highest output
voltage will have to supply all of the current to the load, until its
output drops (due to load regulation drop or current limiting or
thermal limiting) to the output voltage of the second highest one.
And so on. The trick of putting a small resistor in series with the
output makes them share the load. How much resistance? If all the
regulators are within +/-0.2 volts of 5 volts, then the resistor at
the output of each one should drop no more than about 0.4 volts at
the expected shared current level. Say 1 amp per regulator. 0.4
volt/1 amp=~0.4 ohms, rated 1 watt or higher.
-- Tracy
Tim
At 11:39 AM 10/1/2001 -0700, you wrote:
> >well,
> >it might have been for pass transistors on a power supply. There were
> >several 2n3055 in parallel to get 20-30 amps on a 12 volt power supply.
> >Since the transistors are not manufactured 100% exactly the same there are
> >slight differences. If the internal resistances are slightly different then
> >the good transistor carries more than its share of the load and the loafing
> >transistor doesnt do its fair share.
> >
> >The idea was to put a very low resistor in each output terminal connection
> >before tying them all together. It seems that they were 0.33 ohms and rated
> >for about 10 watts. The extra resistance added a small voltage drop but it
> >was worth it to get all transistors to carry their fair share of the load.
> >
> >maybe this applies to regulators also ?
>
>
>That is true of regulators too. The output voltage of one will be a
>little higher than another. The one that has the highest output
>voltage will have to supply all of the current to the load, until its
>output drops (due to load regulation drop or current limiting or
>thermal limiting) to the output voltage of the second highest one.
>And so on. The trick of putting a small resistor in series with the
>output makes them share the load. How much resistance? If all the
>regulators are within +/-0.2 volts of 5 volts, then the resistor at
>the output of each one should drop no more than about 0.4 volts at
>the expected shared current level. Say 1 amp per regulator. 0.4
>volt/1 amp=~0.4 ohms, rated 1 watt or higher.
>
> -- Tracy
>
>
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Timothy Medema
CrystaLite, Inc.
3307 Cedar St. (425) 745-6000 800-666-6065
Everett, WA 98201 Fax: (425) 257-0232
www.crystaliteinc.com
<mailto:timm@c...>timm@c...
The information transmitted is intended only for the person or entity to
whom it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you received
this in error, please contact the sender and delete the material from any
computer.
Tim
At 11:39 AM 10/1/2001 -0700, you wrote:
> >well,
> >it might have been for pass transistors on a power supply. There were
> >several 2n3055 in parallel to get 20-30 amps on a 12 volt power supply.
> >Since the transistors are not manufactured 100% exactly the same there are
> >slight differences. If the internal resistances are slightly different then
> >the good transistor carries more than its share of the load and the loafing
> >transistor doesnt do its fair share.
> >
> >The idea was to put a very low resistor in each output terminal connection
> >before tying them all together. It seems that they were 0.33 ohms and rated
> >for about 10 watts. The extra resistance added a small voltage drop but it
> >was worth it to get all transistors to carry their fair share of the load.
> >
> >maybe this applies to regulators also ?
>
>
>That is true of regulators too. The output voltage of one will be a
>little higher than another. The one that has the highest output
>voltage will have to supply all of the current to the load, until its
>output drops (due to load regulation drop or current limiting or
>thermal limiting) to the output voltage of the second highest one.
>And so on. The trick of putting a small resistor in series with the
>output makes them share the load. How much resistance? If all the
>regulators are within +/-0.2 volts of 5 volts, then the resistor at
>the output of each one should drop no more than about 0.4 volts at
>the expected shared current level. Say 1 amp per regulator. 0.4
>volt/1 amp=~0.4 ohms, rated 1 watt or higher.
>
> -- Tracy
>
>
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Timothy Medema
CrystaLite, Inc.
3307 Cedar St. (425) 745-6000 800-666-6065
Everett, WA 98201 Fax: (425) 257-0232
www.crystaliteinc.com
<mailto:timm@c...>timm@c...
The information transmitted is intended only for the person or entity to
whom it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you received
this in error, please contact the sender and delete the material from any
computer.
the current is not shared equally between the regulators. Take a
worst case, where one regulator outputs 5.2 volts and a second one
outputs 4.8 volts, and each has a 0.4 ohm resistor at its output in
order to "equalize" the load. The one with the 5.2 volt output will
be supplying 1 amp already before the second regulator kicks in.
(0.4 ohm * 1 amp = 0.4 volts, the difference in the outputs). Then,
when the total load on the supply gets up to 3 amps, the 1st
regulator will be supplying 2 amps of that and the second regulator 1
amps worth.
It is unlikely that two voltage regulators will be so far off from 5
volts in opposite directions. But the message is, if possible,
choose voltage regulators from one batch that have the output
voltages close together.
-- Tracy
>Thanks Tracy - much appreciated.
>
>Tim
>
>
>At 11:39 AM 10/1/2001 -0700, you wrote:
>> >well,
>> >it might have been for pass transistors on a power supply. There were
>> >several 2n3055 in parallel to get 20-30 amps on a 12 volt power supply.
>> >Since the transistors are not manufactured 100% exactly the same there are
>> >slight differences. If the internal resistances are slightly different then
>> >the good transistor carries more than its share of the load and the loafing
>> >transistor doesnt do its fair share.
>> >
>> >The idea was to put a very low resistor in each output terminal connection
>> >before tying them all together. It seems that they were 0.33 ohms and rated
>> >for about 10 watts. The extra resistance added a small voltage drop but it
>> >was worth it to get all transistors to carry their fair share of the load.
>> >
>> >maybe this applies to regulators also ?
>>
>>
>>That is true of regulators too. The output voltage of one will be a
>>little higher than another. The one that has the highest output
>>voltage will have to supply all of the current to the load, until its
>>output drops (due to load regulation drop or current limiting or
>>thermal limiting) to the output voltage of the second highest one.
>>And so on. The trick of putting a small resistor in series with the
>>output makes them share the load. How much resistance? If all the
>>regulators are within +/-0.2 volts of 5 volts, then the resistor at
>>the output of each one should drop no more than about 0.4 volts at
>>the expected shared current level. Say 1 amp per regulator. 0.4
>>volt/1 amp=~0.4 ohms, rated 1 watt or higher.
>>
>
> > -- Tracy
alot to this group. What if adjustable voltage regulators were used and
the output voltages were adjusted to match? Would that be close enough and
would it be stable over time?
Tim
At 10:35 AM 10/2/2001 -0700, you wrote:
>A further thought on this: Even with the resistors at the outputs,
>the current is not shared equally between the regulators. Take a
>worst case, where one regulator outputs 5.2 volts and a second one
>outputs 4.8 volts, and each has a 0.4 ohm resistor at its output in
>order to "equalize" the load. The one with the 5.2 volt output will
>be supplying 1 amp already before the second regulator kicks in.
>(0.4 ohm * 1 amp = 0.4 volts, the difference in the outputs). Then,
>when the total load on the supply gets up to 3 amps, the 1st
>regulator will be supplying 2 amps of that and the second regulator 1
>amps worth.
>
>It is unlikely that two voltage regulators will be so far off from 5
>volts in opposite directions. But the message is, if possible,
>choose voltage regulators from one batch that have the output
>voltages close together.
>
> -- Tracy
>
>
> >Thanks Tracy - much appreciated.
> >
> >Tim
> >
> >
> >At 11:39 AM 10/1/2001 -0700, you wrote:
> >> >well,
> >> >it might have been for pass transistors on a power supply. There were
> >> >several 2n3055 in parallel to get 20-30 amps on a 12 volt power supply.
> >> >Since the transistors are not manufactured 100% exactly the same
> there are
> >> >slight differences. If the internal resistances are slightly
> different then
> >> >the good transistor carries more than its share of the load and the
> loafing
> >> >transistor doesnt do its fair share.
> >> >
> >> >The idea was to put a very low resistor in each output terminal
> connection
> >> >before tying them all together. It seems that they were 0.33 ohms and
> rated
> >> >for about 10 watts. The extra resistance added a small voltage drop
> but it
> >> >was worth it to get all transistors to carry their fair share of the
> load.
> >> >
> >> >maybe this applies to regulators also ?
> >>
> >>
> >>That is true of regulators too. The output voltage of one will be a
> >>little higher than another. The one that has the highest output
> >>voltage will have to supply all of the current to the load, until its
> >>output drops (due to load regulation drop or current limiting or
> >>thermal limiting) to the output voltage of the second highest one.
> >>And so on. The trick of putting a small resistor in series with the
> >>output makes them share the load. How much resistance? If all the
> >>regulators are within +/-0.2 volts of 5 volts, then the resistor at
> >>the output of each one should drop no more than about 0.4 volts at
> >>the expected shared current level. Say 1 amp per regulator. 0.4
> >>volt/1 amp=~0.4 ohms, rated 1 watt or higher.
> >>
> >
> > > -- Tracy
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
self protecting via "soft" thermal shutdown, as well as current limiting.
The higher voltage unit would output higher current until it approached
I-limit, where its output voltage would slightly decrease and enable the
other regulators to share. We have seen these beasts survive terrible
abuse.
Regards,
Ray McArthur
Original Message
From: Tracy Allen <tracy@e...>
To: <basicstamps@yahoogroups.com>
Sent: Tuesday, October 02, 2001 1:35 PM
Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
> A further thought on this: Even with the resistors at the outputs,
> the current is not shared equally between the regulators. Take a
> worst case, where one regulator outputs 5.2 volts and a second one
> outputs 4.8 volts, and each has a 0.4 ohm resistor at its output in
> order to "equalize" the load. The one with the 5.2 volt output will
> be supplying 1 amp already before the second regulator kicks in.
> (0.4 ohm * 1 amp = 0.4 volts, the difference in the outputs). Then,
> when the total load on the supply gets up to 3 amps, the 1st
> regulator will be supplying 2 amps of that and the second regulator 1
> amps worth.
>
> It is unlikely that two voltage regulators will be so far off from 5
> volts in opposite directions. But the message is, if possible,
> choose voltage regulators from one batch that have the output
> voltages close together.
>self protecting via "soft" thermal shutdown, as well as current limiting.
>The higher voltage unit would output higher current until it approached
>I-limit, where its output voltage would slightly decrease and enable the
>other regulators to share. We have seen these beasts survive terrible
>abuse.
>
>Regards,
>Ray McArthur
Agreed, these beasts are self protecting and survive terrible abuse.
Still, I think that one of them (when connected directly in parallel
to supply lots of current) might end up operating much too hot for
the long term good of the system. The current limiting is not a
tight parameter, and thermal limiting kicks in when the chip is
operating near its thermal limits. Let's say an LM317T rated at 1.5
amp is set up for 5.1 volts out. Let's say it gets up to full rated
current at 1.5 amps. Up to that point its output voltage has changed
by -20 millivolts at most due to load regulation, and by no more than
about -50 millivolts (1%) due to temperature rise. So If another
LM317 is in parallel with it and is set haphazardly at 4.9 volts out,
it will not contribute anything at all to share the load. To
continue the scenario, let's say that the input-to-output voltage
differential and heat sinking is such that after a few seconds the
first chip starts into thermal limiting. The second regulator will
take just enough of the current to keep the first regulator's silicon
operating at 125 degrees C. That is hot enough to burn a hole in the
circuit board or to lead to other unfortunate consequences.
At 11:26 AM -0700 10/2/01, Timothy Medema wrote:
>What if adjustable voltage regulators were used and
>the output voltages were adjusted to match? Would that be close enough and
>would it be stable over time?
>
>Tim
The situation would be much better if the two regulators were
adjusted initially to less than, say, 20 millivolts difference,
because then the load sharing would kick well before the first chip
reaches limiting. A small amount of resistance (~0.1ohm) in series
between each output and the common load would greatly improve the
sharing. But not too much resistance, because it degrades the load
regulation. I.e., 0.1 ohm * 1 amp = 0.1 volt, while the load
regulation of the LM317 itself over 1 amp load change is less than
0.02 volt. Maybe it's best to use a single TO3 package and a better
heat sink.
Another way to supplement the current from a regulator is by means of
a PNP transistor around the regulator:
power PNP
;
\e c/
;
|
|
| |b |
| R 10ohms |
| | ;
; |
V in ---o--/\/\--o--| |----o----V out
~1ohm | regl |
`
'
|
o
common
When the current required by the load is less than about 0.6 amp, the
PNP transistor is off and all the load current comes through the
regulator. But when the current demand goes above ~0.6 amp, the PNP
transistor turns on and supplies the excess current. It is even
possible to use several PNPs in parallel (with additional 0.1ohm
resistors in the emitters). This circuit does not degrade regulation,
because the transistor is in the feedback loop, but it does degrade
stability, because the transistor adds gain to the loop. The circuit
is not bullet-proof either, because it bypasses the regulator's
current limit sensing. It is easily possible to fry the PNP
transistor! And power PNP transistors are almost as expensive as
full-blown voltage regulators!
-- regards,
Tracy Allen
electronically monitored ecosystems
mailto:tracy@e...
http://www.emesystems.com
>
>
Original Message
>From: Tracy Allen <tracy@e...>
>To: <basicstamps@yahoogroups.com>
>Sent: Tuesday, October 02, 2001 1:35 PM
>Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
>
> > A further thought on this: Even with the resistors at the outputs,
> > the current is not shared equally between the regulators. Take a
> > worst case, where one regulator outputs 5.2 volts and a second one
> > outputs 4.8 volts, and each has a 0.4 ohm resistor at its output in
> > order to "equalize" the load. The one with the 5.2 volt output will
> > be supplying 1 amp already before the second regulator kicks in.
> > (0.4 ohm * 1 amp = 0.4 volts, the difference in the outputs). Then,
> > when the total load on the supply gets up to 3 amps, the 1st
> > regulator will be supplying 2 amps of that and the second regulator 1
> > amps worth.
> >
> > It is unlikely that two voltage regulators will be so far off from 5
> > volts in opposite directions. But the message is, if possible,
> > choose voltage regulators from one batch that have the output
> > voltages close together.
this worked 15 years ago
2N3055 0.1 5W
;
\c e/
/\/\---; all transistors
|
| on the same
| |b | heatsink with
| '
; | the regulator to
| 2N3055 | 0.1 5W | make the overload
o
\c e/
/\/\---o protection
|
| | effective
| |b | |
| '
o |
| 2N3055 | 0.1 5W |
o
\c e/
/\/\---o
|
| |
| |b | |
| '
o |
| ;
; | 100 |
in + ---o---| |---o---/\/\---o--- out, 15A max
| LM317 | |
'
' |
| R2 |
o---/\/\
'
| R1 R1,R2 as required by output V
'---/\/\---;
|
gnd
o
gnd
regards
ECO
Original Message
From: "Tracy Allen" <tracy@e...>
To: <basicstamps@yahoogroups.com>
Sent: Wednesday, October 03, 2001 7:40 AM
Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
> Another way to supplement the current from a regulator is by means of
> a PNP transistor around the regulator:
>
> power PNP
> ;
\e c/
;
> |
|
> | |b |
> | R 10ohms |
> | | ;
; |
> V in ---o--/\/\--o--| |----o----V out
> ~1ohm | regl |
> `
'
> |
>
o
common
>
> When the current required by the load is less than about 0.6 amp, the
> PNP transistor is off and all the load current comes through the
> regulator. But when the current demand goes above ~0.6 amp, the PNP
> transistor turns on and supplies the excess current. It is even
> possible to use several PNPs in parallel (with additional 0.1ohm
> resistors in the emitters). This circuit does not degrade regulation,
> because the transistor is in the feedback loop, but it does degrade
> stability, because the transistor adds gain to the loop. The circuit
> is not bullet-proof either, because it bypasses the regulator's
> current limit sensing. It is easily possible to fry the PNP
> transistor! And power PNP transistors are almost as expensive as
> full-blown voltage regulators!
>
>
> -- regards,
> Tracy Allen
> electronically monitored ecosystems
> mailto:tracy@e...
> http://www.emesystems.com
>
devices. For example the 7800s are protected thermally and also against
output shorcircuit, but I don't think this protections are adequate for
permanent service.
In any case, the junction temperature would be permenntly close to its
limit. This would most probably shorteh the life of the device because of
internal metal migration and other processes. I don't think it's a
good/safe idea.
It may be that other devices could operate like that, I cannot say.
I think that using external boost power transistor(s) is safe and common
practise, as proposed in another contribution in this list
Sorry not to agree...
Regards, Jose Luis
At 17:23 02/10/2001 -0400, you wrote:
>I wouldn't worry about balanced current-sharing since these regulators are
>self protecting via "soft" thermal shutdown, as well as current limiting.
>The higher voltage unit would output higher current until it approached
>I-limit, where its output voltage would slightly decrease and enable the
>other regulators to share. We have seen these beasts survive terrible
>abuse.
>
>Regards,
>Ray McArthur
>
>
Original Message
>From: Tracy Allen <tracy@e...>
>To: <basicstamps@yahoogroups.com>
>Sent: Tuesday, October 02, 2001 1:35 PM
>Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
>
> > A further thought on this: Even with the resistors at the outputs,
> > the current is not shared equally between the regulators. Take a
> > worst case, where one regulator outputs 5.2 volts and a second one
> > outputs 4.8 volts, and each has a 0.4 ohm resistor at its output in
> > order to "equalize" the load. The one with the 5.2 volt output will
> > be supplying 1 amp already before the second regulator kicks in.
> > (0.4 ohm * 1 amp = 0.4 volts, the difference in the outputs). Then,
> > when the total load on the supply gets up to 3 amps, the 1st
> > regulator will be supplying 2 amps of that and the second regulator 1
> > amps worth.
> >
> > It is unlikely that two voltage regulators will be so far off from 5
> > volts in opposite directions. But the message is, if possible,
> > choose voltage regulators from one batch that have the output
> > voltages close together.
>
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
[noparse][[/noparse]Non-text portions of this message have been removed]
This information is just what I'm looking for. I don't know about
the rest of you but I have a very hard time reading the posted schematics -
do either or both of you have them available in another format? I have
Protel's (formerly MicroCode Engineering) Circuit Maker software for
viewing. Maybe you could point me to a web site?
Thanks for the help,
Tim
At 10:36 AM 10/3/2001 +0200, you wrote:
>Tracy
>this worked 15 years ago
>
> 2N3055 0.1 5W
> ;
\c e/
/\/\---; all transistors
> |
| on the same
> | |b | heatsink with
> | '
; | the regulator to
> | 2N3055 | 0.1 5W | make the overload
> o
\c e/
/\/\---o protection
> |
| | effective
> | |b | |
> | '
o |
> | 2N3055 | 0.1 5W |
> o
\c e/
/\/\---o
> |
| |
> | |b | |
> | '
o |
> | ;
; | 100 |
> in + ---o---| |---o---/\/\---o--- out, 15A max
> | LM317 | |
> '
' |
> | R2 |
> o---/\/\
'
> | R1 R1,R2 as required by output V
> '---/\/\---;
> |
> gnd
o
gnd
>
>regards
>ECO
>
Original Message
>From: "Tracy Allen" <tracy@e...>
>To: <basicstamps@yahoogroups.com>
>Sent: Wednesday, October 03, 2001 7:40 AM
>Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
>
> > Another way to supplement the current from a regulator is by means of
> > a PNP transistor around the regulator:
> >
> > power PNP
> > ;
\e c/
;
> > |
|
> > | |b |
> > | R 10ohms |
> > | | ;
; |
> > V in ---o--/\/\--o--| |----o----V out
> > ~1ohm | regl |
> > `
'
> > |
> >
o
common
> >
> > When the current required by the load is less than about 0.6 amp, the
> > PNP transistor is off and all the load current comes through the
> > regulator. But when the current demand goes above ~0.6 amp, the PNP
> > transistor turns on and supplies the excess current. It is even
> > possible to use several PNPs in parallel (with additional 0.1ohm
> > resistors in the emitters). This circuit does not degrade regulation,
> > because the transistor is in the feedback loop, but it does degrade
> > stability, because the transistor adds gain to the loop. The circuit
> > is not bullet-proof either, because it bypasses the regulator's
> > current limit sensing. It is easily possible to fry the PNP
> > transistor! And power PNP transistors are almost as expensive as
> > full-blown voltage regulators!
> >
> >
> > -- regards,
> > Tracy Allen
> > electronically monitored ecosystems
> > mailto:tracy@e...
> > http://www.emesystems.com
> >
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Set your font to courier new and the
schematics reads perfectly.
greetings peter
Oorspronkelijk bericht
Van: Timothy Medema [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=EikagrqbaPvzwAmCvgL5EyMIN2XT4opkES12Bzfv7GzgfZfLNQqoykk4PlMZN1FzY_0eAlFJ9_WWxpxla6IOb9Q]timm@c...[/url
Verzonden: woensdag 3 oktober 2001 15:30
Aan: basicstamps@yahoogroups.com
Onderwerp: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
Tracy and ECO,
This information is just what I'm looking for. I don't know about
the rest of you but I have a very hard time reading the posted schematics -
do either or both of you have them available in another format? I have
Protel's (formerly MicroCode Engineering) Circuit Maker software for
viewing. Maybe you could point me to a web site?
Thanks for the help,
Tim
At 10:36 AM 10/3/2001 +0200, you wrote:
>Tracy
>this worked 15 years ago
>
> 2N3055 0.1 5W
> ;
\c e/
/\/\---; all transistors
> |
| on the same
> | |b | heatsink with
> | '
; | the regulator to
> | 2N3055 | 0.1 5W | make the overload
> o
\c e/
/\/\---o protection
> |
| | effective
> | |b | |
> | '
o |
> | 2N3055 | 0.1 5W |
> o
\c e/
/\/\---o
> |
| |
> | |b | |
> | '
o |
> | ;
; | 100 |
> in + ---o---| |---o---/\/\---o--- out, 15A max
> | LM317 | |
> '
' |
> | R2 |
> o---/\/\
'
> | R1 R1,R2 as required by output V
> '---/\/\---;
> |
> gnd
o
gnd
>
>regards
>ECO
>
Original Message
>From: "Tracy Allen" <tracy@e...>
>To: <basicstamps@yahoogroups.com>
>Sent: Wednesday, October 03, 2001 7:40 AM
>Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
>
> > Another way to supplement the current from a regulator is by means of
> > a PNP transistor around the regulator:
> >
> > power PNP
> > ;
\e c/
;
> > |
|
> > | |b |
> > | R 10ohms |
> > | | ;
; |
> > V in ---o--/\/\--o--| |----o----V out
> > ~1ohm | regl |
> > `
'
> > |
> >
o
common
> >
> > When the current required by the load is less than about 0.6 amp, the
> > PNP transistor is off and all the load current comes through the
> > regulator. But when the current demand goes above ~0.6 amp, the PNP
> > transistor turns on and supplies the excess current. It is even
> > possible to use several PNPs in parallel (with additional 0.1ohm
> > resistors in the emitters). This circuit does not degrade regulation,
> > because the transistor is in the feedback loop, but it does degrade
> > stability, because the transistor adds gain to the loop. The circuit
> > is not bullet-proof either, because it bypasses the regulator's
> > current limit sensing. It is easily possible to fry the PNP
> > transistor! And power PNP transistors are almost as expensive as
> > full-blown voltage regulators!
> >
> >
> > -- regards,
> > Tracy Allen
> > electronically monitored ecosystems
> > mailto:tracy@e...
> > http://www.emesystems.com
> >
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
To UNSUBSCRIBE, just send mail to:
basicstamps-unsubscribe@yahoogroups.com
from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Wow, what a difference that font setting made! You are right, now
the schematics read perfectly.
Very much appreciated,
Tim
At 04:58 PM 10/3/2001 +0200, you wrote:
>Hi,
>
>Set your font to courier new and the
>schematics reads perfectly.
>
>greetings peter
>
>
>
Oorspronkelijk bericht
>Van: Timothy Medema [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=YZLT7EPJ3CJF9N5ngnZKlh8aXvB0ohL2vxQ3Ye4-IEtTS5L_n4CRa0AgBr4NwEMEFJIqx8MMKjr0fhh_uw]timm@c...[/url
>Verzonden: woensdag 3 oktober 2001 15:30
>Aan: basicstamps@yahoogroups.com
>Onderwerp: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
>
>
>Tracy and ECO,
>
> This information is just what I'm looking for. I don't know about
>the rest of you but I have a very hard time reading the posted schematics -
>do either or both of you have them available in another format? I have
>Protel's (formerly MicroCode Engineering) Circuit Maker software for
>viewing. Maybe you could point me to a web site?
>
>Thanks for the help,
>
>Tim
>
>
>At 10:36 AM 10/3/2001 +0200, you wrote:
> >Tracy
> >this worked 15 years ago
> >
> > 2N3055 0.1 5W
> > ;
\c e/
/\/\---; all transistors
> > |
| on the same
> > | |b | heatsink with
> > | '
; | the regulator to
> > | 2N3055 | 0.1 5W | make the overload
> > o
\c e/
/\/\---o protection
> > |
| | effective
> > | |b | |
> > | '
o |
> > | 2N3055 | 0.1 5W |
> > o
\c e/
/\/\---o
> > |
| |
> > | |b | |
> > | '
o |
> > | ;
; | 100 |
> > in + ---o---| |---o---/\/\---o--- out, 15A max
> > | LM317 | |
> > '
' |
> > | R2 |
> > o---/\/\
'
> > | R1 R1,R2 as required by output V
> > '---/\/\---;
> > |
> > gnd
o
gnd
> >
> >regards
> >ECO
> >
Original Message
> >From: "Tracy Allen" <tracy@e...>
> >To: <basicstamps@yahoogroups.com>
> >Sent: Wednesday, October 03, 2001 7:40 AM
> >Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
> >
> >
> > > Another way to supplement the current from a regulator is by means of
> > > a PNP transistor around the regulator:
> > >
> > > power PNP
> > > ;
\e c/
;
> > > |
|
> > > | |b |
> > > | R 10ohms |
> > > | | ;
; |
> > > V in ---o--/\/\--o--| |----o----V out
> > > ~1ohm | regl |
> > > `
'
> > > |
> > >
o
common
> > >
> > > When the current required by the load is less than about 0.6 amp, the
> > > PNP transistor is off and all the load current comes through the
> > > regulator. But when the current demand goes above ~0.6 amp, the PNP
> > > transistor turns on and supplies the excess current. It is even
> > > possible to use several PNPs in parallel (with additional 0.1ohm
> > > resistors in the emitters). This circuit does not degrade regulation,
> > > because the transistor is in the feedback loop, but it does degrade
> > > stability, because the transistor adds gain to the loop. The circuit
> > > is not bullet-proof either, because it bypasses the regulator's
> > > current limit sensing. It is easily possible to fry the PNP
> > > transistor! And power PNP transistors are almost as expensive as
> > > full-blown voltage regulators!
> > >
> > >
> > > -- regards,
> > > Tracy Allen
> > > electronically monitored ecosystems
> > > mailto:tracy@e...
> > > http://www.emesystems.com
> > >
> >
> >
> >To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@yahoogroups.com
> >from the same email address that you subscribed. Text in the Subject and
> >Body of the message will be ignored.
> >
> >
> >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>
>
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and
>Body of the message will be ignored.
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
breadboard. The booster transistor is a much better approach.
Ray McArthur
Original Message
From: Tracy Allen <tracy@e...>
To: <basicstamps@yahoogroups.com>
Sent: Wednesday, October 03, 2001 1:40 AM
Subject: Re: [noparse][[/noparse]basicstamps] Parallel Voltage Regulators....
>
> Agreed, these beasts are self protecting and survive terrible abuse.
> Still, I think that one of them (when connected directly in parallel
> to supply lots of current) might end up operating much too hot for
> the long term good of the system. The current limiting is not a
> tight parameter, and thermal limiting kicks in when the chip is
> operating near its thermal limits. Let's say an LM317T rated at 1.5
> amp is set up for 5.1 volts out. Let's say it gets up to full rated
> current at 1.5 amps. Up to that point its output voltage has changed
> by -20 millivolts at most due to load regulation, and by no more than
> about -50 millivolts (1%) due to temperature rise. So If another
> LM317 is in parallel with it and is set haphazardly at 4.9 volts out,
> it will not contribute anything at all to share the load. To
> continue the scenario, let's say that the input-to-output voltage
> differential and heat sinking is such that after a few seconds the
> first chip starts into thermal limiting. The second regulator will
> take just enough of the current to keep the first regulator's silicon
> operating at 125 degrees C. That is hot enough to burn a hole in the
> circuit board or to lead to other unfortunate consequences.