Maker Pro
Maker Pro

ESR readings

M

Mike

Jan 1, 1970
0
Im working on infocus power supplys, is an ESR of 1 to 1.5 ohm alright for
100uf caps at 25volts? they are in the primary circuit of a power supply.
 
A

Arfa Daily

Jan 1, 1970
0
Mike said:
Im working on infocus power supplys, is an ESR of 1 to 1.5 ohm alright for
100uf caps at 25volts? they are in the primary circuit of a power supply.
No, that's too high. Expect to see around 0.3 ohms for a 'standard' cap,
less than that for a low ESR type. However, I doubt that 1 to 1.5 ohms is
bad enough yet to be stopping the supply from working, but replace them
anyway.

Arfa
 
M

Mike

Jan 1, 1970
0
most of the supplys work, but the boostup converter is REAL weak at startup.
they "grunt" before the bulb will start to grow brighter. normal ones dont
do that, and I changed the caps, still "grunts" so im assuming worn out
MOSFET in the boostup converter.
 
M

Mike

Jan 1, 1970
0
well 0.47 caps, 1uf caps, and 2.2 uf caps read about 5. at 50khz, new and
old. so im assuming thats how it reads.


all 107 power supplys I have here, all 3 100uf caps are reading near 1 ohm,
some a little over.


my new 100uf reads right at 0. (analog meter).
 
M

Mike

Jan 1, 1970
0
a more detailed explanation, it will strike the lamp, and the lamp will have
a really dim arc. for a few seconds. then youll hear a small reeeeeek grunt
sound like weak capacitors would cause in a normal power supply, but after
that sound, the bulb snaps really bright and starts growing brighter like a
normal lamp.


So i went right to the capacitors, they all read near 1 ohm or a little over
for the 100uf rubycons. all my new ones read right at 0.

the main supply output filters read right near 0. so they are ok.

the rest is the mylar box caps. they are high enough value to be picked up
by the meter, none or are open or shorted so im assuming worn out
semiconductors from the heat. the astec supplys get really hot really fast.
the magnateks dont do that.
 
A

Arfa Daily

Jan 1, 1970
0
Mike said:
well 0.47 caps, 1uf caps, and 2.2 uf caps read about 5. at 50khz, new and
old. so im assuming thats how it reads.


all 107 power supplys I have here, all 3 100uf caps are reading near 1
ohm, some a little over.


my new 100uf reads right at 0. (analog meter).
I'm not really following that. What relevance does the readings of those
other values have, in respect of the value you were asking about ? Your
reading of " 0. " on an " analogue meter " is also confusing me. Zero is a
short circuit. What sort of analogue meter are we talking here - ESR or ohms
? I have just checked a brand new out-of-the-drawer bog-standard 100uF @
25v, and on my Bob Parker meter, and it goes 0.3 ohms, which is in agreement
with the front panel legend.

As I said before, 1 ohm plus for this value, is a little high. If I were
repairing a SMPS, and found a cap of that value reading thus, I would feel
inclined to replace it, as being " on the way out ", not because I would be
particularly confident of it causing a problem. Although electrolytic caps
are probably the single biggest source of trouble in switchers, there is a
designed-in tolerance in most reputable designs, such that caps have to be
significantly out of spec, before they start causing real problems. ( 5 or 6
ohms possibly for that value cap )

As far as your other belief ( further down the thread ) that " worn out "
semiconductors - MOSFETs - are to blame, I think that the chances of that
lie somewhere between slim and zero. Whilst I have had bipolar transistors
that have gone noisy, and some that have gone low gain, and more than a few
that have gone leaky over the years, by far the most common failure mode for
semiconductors, and particularly power MOSFETs, is catastrophic junction
failure e.g. short circuit. I cannot think of a failure mode that might be
considered as " worn out ".

Depending on the topology employed in these supplies - self oscillating
primary, IC oscillator and control primary etc - if caps are not to blame
for the sluggish startup, then I would be inclined to start looking for
resistors that have gone high. Possibly a startup resistor on the primary
side, or a current sense resistor on the secondary side. Good luck with
them, and do be careful to observe the highly necessary safety precautions
for working on line powered switchers.

Arfa
 
Mike said:
a more detailed explanation, it will strike the lamp, and the lamp will have
a really dim arc. for a few seconds. then youll hear a small reeeeeek grunt
sound like weak capacitors would cause in a normal power supply, but after
that sound, the bulb snaps really bright and starts growing brighter like a
normal lamp.

sounds normal.


NT
 
M

Mike

Jan 1, 1970
0
LOL i doubt it. i have a good supply that starts up right away.


im gonna take the other persons advice, i think the resistors are going out
of tolerance on the power factor controler. now i have to find a magnafying
glass to read the number on these SMD resistors.
 
M

Mike

Jan 1, 1970
0
I dont have a bob parker or a dick smith meter, I just got an average MCM
tenma run of the mill analog ESR meter. its hard to read down to the decimal
points, but it gives you a general idea at least. :) I dont have an ESR
chart to follow, im new to this ESR stuff. all my power supplies that have
100uf caps are all reading right on 1ohm. some read 2 ohms, and i know thats
bad. I just didnt know what was good and what was bad, again i dont have an
ESR chart and im new to ESR.
 
M

Mike

Jan 1, 1970
0
as far as sluggish startup, there is 2 parts to this supply, well actually
3, but 2 are most important.

there is a VIPER100A switcher circuit that runs the transformer that
produces the 5, 12, 3.3, -8.5 volts DC. That works fine.


its the 120 to 400v boostup thats being sluggish to start. as soon as the
bulb starts, a normal power supply the boost would startup instantly.

these few I got here are a little sluggish than others. some boost
converters dont start at all, and im beginning to go toward your resistor
theory. becuase the mosfet, and the power factor controlle rIC that runs the
boostup test fine.

there is only 1 cap in the boostup converter besides the tank cap. replaced
the cap in the feedback return, and didnt help much. its a little more
regulated now :) i guess thats a plus.
 
Top