9 Jun
2007
9 Jun
'07
7:36 p.m.
While conversing with another list member about some old components, I was
reminded of hearing somewhere that there was a nasty failure mode of
the LM309K 5V regulator (it being the (or one of the) first integrated 5V
regs) which might make it somewhat undesirable for current use.
I don't want to declare that there is such a problem - it's a vague
second-hand rumour to me - so can anyone confirm or elaborate on this?
I've never personally run across problems with them..
A 76 Moto databook indicates it has those nice things like short-circuit
protection and thermal-overload protection.. ..another device perhaps?
9 Jun
9 Jun
8:23 p.m.
A number of three terminal regulators exhibit problems if capacitances
they see on their input and/or output are wrong.
Brent Hilpert wrote:
While conversing with another list member about some
old components, I was
reminded of hearing somewhere that there was a nasty failure mode of
the LM309K 5V regulator (it being the (or one of the) first integrated 5V
regs) which might make it somewhat undesirable for current use.
I don't want to declare that there is such a problem - it's a vague
second-hand rumour to me - so can anyone confirm or elaborate on this?
I've never personally run across problems with them..
A 76 Moto databook indicates it has those nice things like short-circuit
protection and thermal-overload protection.. ..another device perhaps?
10 Jun
10 Jun
1:23 a.m.
From: Brent Hilpert <hilpert at cs.ubc.ca>
While conversing with another list member about some old components, I was
reminded of hearing somewhere that there was a nasty failure mode of
the LM309K 5V regulator (it being the (or one of the) first integrated 5V
regs) which might make it somewhat undesirable for current use.
I don't want to declare that there is such a problem - it's a vague
second-hand rumour to me - so can anyone confirm or elaborate on this?
I've never personally run across problems with them..
A 76 Moto databook indicates it has those nice things like short-circuit
protection and thermal-overload protection.. ..another device perhaps?
Hi
Many regulators could not handle significant back voltage. The problem
would often occur if the input had a small capacitor relative to the
output and there was a bleader or other load on the input source.
This was common problem with many CMOS circuits.
Another problem was input over voltage. Often if ther was a transformer
without an input filter capacitor, when the power was turned off, the
inductive spike would wipe out the regulator. It seems like negative
rail regulators were most suseptible to this.
Dwight
_________________________________________________________________
Need a break? Find your escape route with Live Search Maps.
http://maps.live.com/default.aspx?ss=Restaurants~Hotels~Amusement%20Park&am…
4:40 a.m.
Van: cctalk-bounces at classiccmp.org namens dwight
elvey
Verzonden: zo 10-06-2007 07:23
Aan: cctalk at classiccmp.org
Onderwerp: RE: LM309K regulator
I would like to hear more about this! I am about to build myself a small
power supply for -16V. using an LM337T. I know that if the output capacitor
is larger than the input filter cap you better put a diode connected between
the output and input pin, and that for HF you should put small caps (say
10 to 100 nF) with short leads from input to Gnd and output to Gnd.
Never heard about that switch-off spike from the transformer.
Since that -16V. is the power supply for my RACAL RA.366 panoramic adapter
I'd hate to see that unit get damaged ... would a VDR across the secundary
of the transformer help?
thanks,
- Henk, PA8PDP
This message and attachment(s) are intended solely for use by the addressee and may
contain information that is privileged, confidential or otherwise exempt from disclosure
under applicable law.
If you are not the intended recipient or agent thereof responsible for delivering this
message to the intended recipient, you are hereby notified that any dissemination,
distribution or copying of this communication is strictly prohibited.
If you have received this communication in error, please notify the sender immediately by
telephone and with a 'reply' message.
Thank you for your co-operation.
From: Brent Hilpert <hilpert at cs.ubc.ca>
While conversing with another list member about some old components, I
was reminded of hearing somewhere that there was a nasty failure mode of
the LM309K 5V regulator (it being the (or one of the) first integrated 5V
regs) which might make it somewhat undesirable for current use.
I don't want to declare that there is such a problem - it's a vague
second-hand rumour to me - so can anyone confirm or elaborate on this?
I've never personally run across problems with them..
A 76 Moto databook indicates it has those nice things like short-circuit
protection and thermal-overload protection.. ..another device perhaps?
Hi
Many regulators could not handle significant back voltage. The problem
would often occur if the input had a small capacitor relative to the
output and there was a bleader or other load on the input source.
This was common problem with many CMOS circuits.
Another problem was input over voltage. Often if ther was a transformer
without an input filter capacitor, when the power was turned off, the
inductive spike would wipe out the regulator. It seems like negative
rail regulators were most suseptible to this.
Dwight
12:33 p.m.
From: "Gooijen, Henk" <henk.gooijen at
oce.com>
I would like to hear more about this! I am about to build myself a small
power supply for -16V. using an LM337T. I know that if the output capacitor
is larger than the input filter cap you better put a diode connected
between
the output and input pin, and that for HF you should put small caps (say
10 to 100 nF) with short leads from input to Gnd and output to Gnd.
Never heard about that switch-off spike from the transformer.
Since that -16V. is the power supply for my RACAL RA.366 panoramic adapter
I'd hate to see that unit get damaged ... would a VDR across the secundary
of the transformer help?
thanks,
- Henk, PA8PDP
Hi Henk
Usually the filter cap on the rectifier is enough to safely absorb the
energy
of the transformer. Still, it is something that you need to check and
actually
verify. If the range of line voltages that the unit is to run on put the
input
voltage near the maximum of the rating, it needs to be checked.
Where I personally saw this effect was in a unit that had 5 of the modular
power supplies ( linear w/ transformers ) all wired in parallel. When the
switch was turn on and off, the stored energy in the cores needs to
go sumplace, just like the coil in an ignition system.
What makes it worse is something like a zero cross switch which would be
worse
as that is when the maximum energy is in the cores magnetic field.
We were not able to get the manufacture of the modular power supplies
to solve the problem fast enough so we shipped the machine but put
those varistors on the inputs to clamp the voltage swing. We learned
later that sufficient sized input caps would also work. For these, one
had to make sure that it wasn't too small as this would actually enhance
the spike, as the capacitor does in an ignition system does, buy reducing
the arcing of the switch contacts.
The diode, as you mentions, was the typical way to solve the back
voltage problem. It is a good idea in most any case.
Dwight
_________________________________________________________________
Don?t miss your chance to WIN $10,000 and other great prizes from Microsoft
Office Live http://clk.atdmt.com/MRT/go/aub0540003042mrt/direct/01/
Van: cctalk-bounces at classiccmp.org namens
dwight elvey
Verzonden: zo 10-06-2007 07:23
Aan: cctalk at classiccmp.org
Onderwerp: RE: LM309K regulator
>From: Brent Hilpert <hilpert at cs.ubc.ca>
>
>While conversing with another list member about some old components, I
>was reminded of hearing somewhere that there was a nasty failure mode of
>the LM309K 5V regulator (it being the (or one of the) first integrated
5V
regs)
which might make it somewhat undesirable for current use.
I don't want to declare that there is such a problem - it's a vague
second-hand rumour to me - so can anyone confirm or elaborate on this?
I've never personally run across problems with them..
A 76 Moto databook indicates it has those nice things like short-circuit
protection and thermal-overload protection.. ..another device perhaps?
Hi
Many regulators could not handle significant back voltage. The problem
would often occur if the input had a small capacitor relative to the
output and there was a bleader or other load on the input source.
This was common problem with many CMOS circuits.
Another problem was input over voltage. Often if ther was a transformer
without an input filter capacitor, when the power was turned off, the
inductive spike would wipe out the regulator. It seems like negative
rail regulators were most suseptible to this.
Dwight
11 Jun
11 Jun
3:25 a.m.
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] On Behalf Of dwight elvey
Sent: zondag 10 juni 2007 18:34
To: cctalk at classiccmp.org
Subject: RE: LM309K regulator
it's a
vague
the output
Thanks Dwight,
I will use a large input capacitor and a VDR (varistor) just
to be on the safe side. But I will check with a dummy load
on the PSU, as I rather not start repairs on the RA.366 ...
- Henk, PA8PDP.
This message and attachment(s) are intended solely for use by the addressee and may
contain information that is privileged, confidential or otherwise exempt from disclosure
under applicable law.
If you are not the intended recipient or agent thereof responsible for delivering this
message to the intended recipient, you are hereby notified that any dissemination,
distribution or copying of this communication is strictly prohibited.
If you have received this communication in error, please notify the sender immediately by
telephone and with a 'reply' message.
Thank you for your co-operation.
From: "Gooijen, Henk" <henk.gooijen
at oce.com>
>Van: cctalk-bounces at classiccmp.org namens dwight elvey
>Verzonden: zo 10-06-2007 07:23
>Aan: cctalk at classiccmp.org
>Onderwerp: RE: LM309K regulator
>
>>From: Brent Hilpert <hilpert at cs.ubc.ca>
>>
>>While conversing with another list member about some old
components,
>>I was reminded of hearing somewhere that
there was a
nasty failure
>>mode of the LM309K 5V regulator (it being
the (or one of
the) first
>integrated
5V
>>regs) which might make it somewhat undesirable for current use.
>>
>>I don't want to declare that there is such a problem - >>second-hand rumour to me - so can anyone
confirm or
elaborate on this?
>>I've never personally run across
problems with them..
>>A 76 Moto databook indicates it has those nice things like
>>short-circuit protection and thermal-overload
protection.. ..another
device perhaps?
>
>
>Hi
>Many regulators could not handle significant back voltage. The
>problem would often occur if the input had a small
capacitor relative
>to the output and there was a bleader or
other load on the
input source.
This was
common problem with many CMOS circuits.
Another problem was input over voltage. Often if ther was a
transformer without an input filter capacitor, when the power was
turned off, the inductive spike would wipe out the regulator. It
seems like negative rail regulators were most suseptible to this.
Dwight
I would like to hear more about this! I am about to build myself a
small power supply for -16V. using an LM337T. I know that if capacitor is larger than the input filter cap you
better put a diode
connected between the output and input pin, and that for HF
you should
put small caps (say 10 to 100 nF) with short leads
from input to Gnd
and output to Gnd.
Never heard about that switch-off spike from the transformer.
Since that -16V. is the power supply for my RACAL RA.366 panoramic
adapter I'd hate to see that unit get damaged ... would a VDR across
the secundary of the transformer help?
thanks,
- Henk, PA8PDP
Hi Henk
Usually the filter cap on the rectifier is enough to safely
absorb the energy of the transformer. Still, it is something
that you need to check and actually verify. If the range of
line voltages that the unit is to run on put the input
voltage near the maximum of the rating, it needs to be checked.
Where I personally saw this effect was in a unit that had 5
of the modular power supplies ( linear w/ transformers ) all
wired in parallel. When the switch was turn on and off, the
stored energy in the cores needs to go sumplace, just like
the coil in an ignition system.
What makes it worse is something like a zero cross switch
which would be worse as that is when the maximum energy is in
the cores magnetic field.
We were not able to get the manufacture of the modular power
supplies to solve the problem fast enough so we shipped the
machine but put those varistors on the inputs to clamp the
voltage swing. We learned later that sufficient sized input
caps would also work. For these, one had to make sure that it
wasn't too small as this would actually enhance the spike, as
the capacitor does in an ignition system does, buy reducing
the arcing of the switch contacts.
The diode, as you mentions, was the typical way to solve the
back voltage problem. It is a good idea in most any case.
Dwight
10 Jun
10 Jun
1:22 p.m.
Steve Stutman wrote:
A number of three terminal regulators exhibit problems
if capacitances
they see on their input and/or output are wrong.
dwight elvey wrote:
Many regulators could not handle significant back
voltage. The problem
would often occur if the input had a small capacitor relative to the
output and there was a bleader or other load on the input source.
This was common problem with many CMOS circuits.
Another problem was input over voltage. Often if ther was a transformer
without an input filter capacitor, when the power was turned off, the
inductive spike would wipe out the regulator. It seems like negative
rail regulators were most suseptible to this.
Yes, I was aware of these issues, they're covered in the databooks for the
later regulators in the fine print.
Guessing, but perhaps before these issues were commonly realised or their
solutions incorporated, problems showed up more frequently, and they would
show up in earlier systems which were more likely to be using the 309, and so
the 309 unfairly got a bad rap.
BTW: The earliest reference I found for the 309 was Feb 1971 (Nat Semi Linear
Application Note). Mentioned as one of the motivations for it's development or
uses is distributed/on-card regulation, the technique the Altair/S100 bus
would later take up with.
6229
days inactive
6231
days old
6 comments
4 participants
participants (4)
-
dkelvey＠hotmail.com -
henk.gooijen＠oce.com -
hilpert＠cs.ubc.ca -
steve＠radiorobots.com