Failure of 'startup" resistors in SMPS
Franc Zabkar
resistor between the positive terminal of the mains filter cap and the
base of the chopper transistor or regulator.
In my experience these resistors always have wattage ratings far in
excess of that experienced in circuit. Yet they fail by quietly going
O/C.
What is the reason for these failures? Is it related to their
composition, ie carbon as opposed to metal film?
-- Franc Zabkar
Please remove one 'g' from my address when replying by email.
Steve Bell
Franc Zabkar wrote in message <36512dd...@news.bigpond.com>...
>I find that a common failure mode of a SMPS is an O/C 'startup'
>resistor between the positive terminal of the mains filter cap and the
>base of the chopper transistor or regulator.
>
>In my experience these resistors always have wattage ratings far in
>excess of that experienced in circuit. Yet they fail by quietly going
>O/C.
>
>What is the reason for these failures? Is it related to their
>composition, ie carbon as opposed to metal film?
>
They are stressed at startup, and from the look of some resistors, they
probably aren't of sufficient voltage rating - something often overlooked. I
always use small PR02 types, small package, 2W and 500V working
Steve Bell
Sam Goldwasser
A factor of two isn't much of a margine considering the contanst operation
AND heat of surrounding components.
For example, R502(?) in those Sylvania TVs that fails is 47 K but is
dissipating almost 1/2 W. Even if it is a 1 W resistor (I don't think it is
any larger), that is too close in my book.
--- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Mirror Info: http://www.repairfaq.org/REPAIR/F_mirror.html
| Sam's Laser FAQ: http://www.misty.com/~don/lasersam.html
In article <36512dd...@news.bigpond.com> Franc...@biggpond.com (Franc Zabkar) writes:
I find that a common failure mode of a SMPS is an O/C 'startup'
resistor between the positive terminal of the mains filter cap and the
base of the chopper transistor or regulator.
In my experience these resistors always have wattage ratings far in
excess of that experienced in circuit. Yet they fail by quietly going
O/C.
What is the reason for these failures? Is it related to their
composition, ie carbon as opposed to metal film?
-- Franc Zabkar
Sam Goldwasser
all - just run too near their ratings in a hot environment as well!
--- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Mirror Info: http://www.repairfaq.org/REPAIR/F_mirror.html
| Sam's Laser FAQ: http://www.misty.com/~don/lasersam.html
In article <72ujpr$5iv$1...@newsreader3.core.theplanet.net> "Steve Bell" <ser...@bell-electronics.freeserve.co.uk> writes:
Franc Zabkar wrote in message <36512dd...@news.bigpond.com>...
>I find that a common failure mode of a SMPS is an O/C 'startup'
>resistor between the positive terminal of the mains filter cap and the
>base of the chopper transistor or regulator.
>
>In my experience these resistors always have wattage ratings far in
>excess of that experienced in circuit. Yet they fail by quietly going
>O/C.
>
>What is the reason for these failures? Is it related to their
>composition, ie carbon as opposed to metal film?
>
They are stressed at startup, and from the look of some resistors, they
bob parker
>Are the wattage ratings really THAT much in excess of the normal dissipation?
>A factor of two isn't much of a margine considering the contanst operation
>AND heat of surrounding components.
>
>For example, R502(?) in those Sylvania TVs that fails is 47 K but is
>dissipating almost 1/2 W. Even if it is a 1 W resistor (I don't think it is
>any larger), that is too close in my book.
>
> --- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
> Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
> +Lasers | Mirror Info: http://www.repairfaq.org/REPAIR/F_mirror.html
> | Sam's Laser FAQ: http://www.misty.com/~don/lasersam.html
>
>In article <36512dd...@news.bigpond.com> Franc...@biggpond.com (Franc Zabkar) writes:
>
> I find that a common failure mode of a SMPS is an O/C 'startup'
> resistor between the positive terminal of the mains filter cap and the
> base of the chopper transistor or regulator.
>
> In my experience these resistors always have wattage ratings far in
> excess of that experienced in circuit. Yet they fail by quietly going
> O/C.
>
> What is the reason for these failures? Is it related to their
> composition, ie carbon as opposed to metal film?
>
> -- Franc Zabkar
>
> Please remove one 'g' from my address when replying by email.
Hi,
From my observations, even when the resistors are dissipating well
below their ratings, they still often go O/C for no apparent reason
without any signs of having overheated. I'm kinda suspicious that the
constant high DC voltage has something to do with it.... maybe some
kind of electrolytic corrosion going on beneath the exterior?
I've always replaced them with generously rated metal film ones and
haven't (yet) seen one of those fail in the same way.
bob parker- electronics technician, sydney australia
thought for the month: it's hard to write software in between writing e-mails!
e-mail: bo...@nlc.net.au
home page: http://www.nlc.net.au/~bobp/
phone/fax: +61 2 9587 8148
Asimov
Re: Failure of 'startup" resistors in SMPS (18 Nov 98 10:25:47)
sa> Are the wattage ratings really THAT much in excess of the normal
sa> dissipation? A factor of two isn't much of a margine considering the
sa> contanst operation AND heat of surrounding components.
sa> For example, R502(?) in those Sylvania TVs that fails is 47 K but is
sa> dissipating almost 1/2 W. Even if it is a 1 W resistor (I don't think
sa> it is any larger), that is too close in my book.
Rule of thumb a resistor is fine up to 75% of its power rating, so 50% is
well in excess of this, otherwise why bother with ratings. However the
voltage rating and how it relates to powerline spikes is another matter.
Read my previous reply describing the capacitive divider effect and spikes.
Is the resistor body mounted in contact with the PCB? When replacing these
try mounting them standing off by leaving the leads a little longer.
... [] <- Please write your complaint legibly in that box.
--
| Return Address: mike...@juxta.mnet.pubnix.ten
| Standard disclaimer: The views of this user are strictly her/his own.
| From addresses mangled solely to block spamming.
| Apologies to those wishing to respond, correct suffix with .net
| Signature auto-added at gateway.
Asimov
Failure of 'startup" resistors in SMPS (18 Nov 98 07:44:53)
Fr> I find that a common failure mode of a SMPS is an O/C 'startup'
Fr> resistor between the positive terminal of the mains filter cap and the
Fr> base of the chopper transistor or regulator.
Fr> In my experience these resistors always have wattage ratings far in
Fr> excess of that experienced in circuit. Yet they fail by quietly going
Fr> O/C.
Fr> What is the reason for these failures? Is it related to their
Fr> composition, ie carbon as opposed to metal film?
Fr> -- Franc Zabkar
Fr> Please remove one 'g' from my address when replying by email.
It's hard to say why these fail so much. I can only make a few
guesses. Even though a resistor may have a big power rating, it can
still have a relatively low breakdown voltage. Offhand I think the
typical resistor voltage rating is around 500 (?) volts. It could be
spikes on the powerline zapping them and they can easily get by the
filter electros as these have a relatively high inductance.
Related to spike energy is capacitive effects. Imagine if you will
that the resistor's body is in contact with the circuit ground. This
is good for power dissipation but might be a bad idea because of the
physical capacitive divider which is formed.
The reason it's a bad idea is because a spike is composed of very high
frequencies and will find less impedance through 5pF's than 100K.
There is some supporting evidence to this notion because if you
examine the area carefully you'll often find a slight trace of soot
left on the circuit board and inbetween the resistor body. In light of
this it would probably be better to replace start-up resistors
standing off the circuit board by leaving the legs longer. Just an
idea to explain the mystery.
pama...@letterbox.com
What I normally do with this kind of problem with SMPS from TVs to PC
Power Supply is replacing it with a series equivalent of the resistor,
this way the resistor can tolerate greater potential that arises from
abnormal conditions. I don't understand why consumer electronics
designers don't do their homework. These methods have been used for HV
ckts. for decades... Using expensive parts isn't the solution to this
problem, it is proper design techniques.
>
Rudolf Ladyzhenskii
Say, resistor cost 1 cent.
You are major PC power suplly manufacturer and making 1 000 000 units per
year. By putting two resistors in, you are loosing 1 000 000 cents or 10 000
dollars. Simple math. Alyhough, those resistors usually fail in a few years
time, when unit is out of warranty and consumers will have either to repair
it or get a new one (which generates more work for the manufacturer).
Working as a design engineer, I sometimes forced to make unit "cheaper".
This leads to sacrfising some of the protection in the unit or degrading its
performance. Electronics design engineers do they homework, but there is als
a marketing people who do not care. Their job is to get maximum profit.
Rudolf
pama...@letterbox.com wrote in message <365371A7...@letterbox.com>...
Steve Bell
the service industry its a good thing, those jobs compensate for the pigs
that can take hours. For the user, thats different.
The 'standard' resistors we use here for general use are MRS25 type, 0.6W 1%
250V. Others of similar wattage and composition vary between 200V and 350V.
We often see cheap low quality resistors, often 2 in series in various power
supplies. Metal film 2W 500V are a good replacement. Any power resistor
stood off the board is a good thing. Many we change are stood up on ceramic
standoffs.
What is a mystery to me, is why the typical 2W startup resistor is often
encapsulated in heat shrink tubing? This must make it more prone to failure,
or am I missing something? The starup cap is also often covered the same
way. I can only assume that is an attempt to lengthen its life, protecting
it from the heat generated by nearby components or the inverter FET
heatsink?
Covering other devices, such as MOV Varistors is logical, as when they short
they splat over everything nearby.
Steve Bell
Franc Zabkar
to keyboard and composed:
>What I normally do with this kind of problem with SMPS from TVs to PC
>Power Supply is replacing it with a series equivalent of the resistor,
>this way the resistor can tolerate greater potential that arises from
>abnormal conditions. I don't understand why consumer electronics
>designers don't do their homework. These methods have been used for HV
>ckts. for decades... Using expensive parts isn't the solution to this
>problem, it is proper design techniques.
Funny you should say that. My post was prompted by a failure in a CTX
monitor where the nominal 820K startup resistor was made up of a 390K
and a 420K in series. The 420K had failed.
-- Franc Zabkar
Franc Zabkar
composed:
>Are the wattage ratings really THAT much in excess of the normal dissipation?
>A factor of two isn't much of a margine considering the contanst operation
>AND heat of surrounding components.
I just repaired a CTX monitor whose startup resistor consists of a
390K and a 420K in series. The DC voltage is 340V.
Total power dissipation ~ 350 x 350 / 800 = 150 mW
Therefore each resistor dissipates 75 mW and has a voltage drop of
170V.
Sam Goldwasser
1/8 W: 200 V
1/4 W: 250 V
1/2 W: 350 V
Surge ratings about twice these.
--- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Mirror Info: http://www.repairfaq.org/REPAIR/F_mirror.html
| Sam's Laser FAQ: http://www.misty.com/~don/lasersam.html
Steve Bell
>
>I just repaired a CTX monitor whose startup resistor consists of a
>390K and a 420K in series. The DC voltage is 340V.
>
>Total power dissipation ~ 350 x 350 / 800 = 150 mW
>
>Therefore each resistor dissipates 75 mW and has a voltage drop of
>170V.
>
>-- Franc Zabkar
That would definitely be voltage related problem. I see 2 types of startup
circuitry, 1 to startup IC's such as UC3842's, the other like your CTX, a
much higher value lower power (and usualy low quality) resistor/s, to give
the inverter transistor base an initial kick to get the circuitry going. I
use some metal film restors rated at 2.5KV, usually for focus and G2 related
area's, but cannot buy under 1.2M.
Steve Bell
PWhite4
in parallel for the start resistors in smps circuits. I can't remember the
exact term, worm effect or something, it just isn't coming to me right now.
He explained that the frequencies that the supplies are working at causes RF to
creep over these resistors. That would help another poster's explination of
possible circuit board capacitance between the resistors and the board causeing
problem. Anyway this effect damages the resistor.
PW
Steve Bell
PWhite4 wrote in message <19981119224700...@ng150.aol.com>...
The most resistors I've ever seen were in Panasonic laser printer power
supplies, very tiny and about 5 or 6 in series, I used to think that odd.
Steve Bell
PWhite4
came to me as I was scanning the posts tonight. By putting larger value
resistors in parallel you increase the surface area of the resistors to
minimize the "skin effect". It must work because I don't recall seeing a
problem in the sony TV power supplies of this nature. To bad they have the
other problems with shorting out the regulators for various reasons.
PW
Tom MacIntyre
>The term Sony used is "skin effect" for the RF creeping over the resistors. It
>came to me as I was scanning the posts tonight. By putting larger value
>resistors in parallel you increase the surface area of the resistors to
>minimize the "skin effect". It must work because I don't recall seeing a
>problem in the sony TV power supplies of this nature. To bad they have the
>other problems with shorting out the regulators for various reasons.
>
>PW
>
Maybe they should be making/using litz wirewound resistors.
Tom