Dell 1800FP no 5v standby.

[bammbammfran]

Re: Dell 1800FP no 5v standby.

if the mosfet and controller chip are out, you should be able to check them in circuit
if one reads incorrectly , then remove that one and retest it out of circuit

[budm]

Re: Dell 1800FP no 5v standby.

You got the got the RD network for the Gate correctly, good job. As you can also, this power supply has not clamping Zener Diode ZD902 (22V) to clamp down the spike on the VCC when the power supply first starts up which kill the IC also (I was real lucky that I never have the MOSFET shorted out Gate to Drain that destroys the IC), the same power supply used in GATEWAY has that Zener installed, I guess Dell cheaps out on the part.

[triplefour]

Re: Dell 1800FP no 5v standby.

can i add the zener diode myself? also how is a zener diode different from a regular one?

[triplefour]

Re: Dell 1800FP no 5v standby.

ok testing the power mosfet (still on the board)
with black probe on DRAIN (middle pin)
i get 8.25k ohms resistance to source
and 61.5k ohms resistance to the gate
resistance between the side pins (source and gate) doesnt change with probes reversed and is 9.89kohms either way

with probes reversed, red probe on drain
resistance to source is infinite
resistance to gate is infinite


oh and also something weird happens and ive seen this before and still dont understand it.
with black probe on DRAIN (middle pin)
i get 8.25k ohms resistance to source
this measurement i took when the meter was in 20k ohms scale.
if i change the scale i get totally different numbers. i would expect the numbers to be the same but just the decimal point changes.
for instance if i have it on 200 ohms scale i get infinite (makes sense)
on 2000 ohms scale it reads 497
on 20k ohms it reads 8.25k
on 200k ohms it reads 47.3k
and on 2000k ohms it reads 348k

can someone please explain what is happening here? why does changing scale change the measurement?

[triplefour]

Re: Dell 1800FP no 5v standby.

with the mosfet removed i am now testing the diodes and resistors that were connected to the gate pin and i find that for d907 i get 95.2ohms resistance going either way. does this mean the diode has failed? or is it still "in circuit" ? only one way to find out i guess...

[triplefour]

Re: Dell 1800FP no 5v standby.

with the diode out i get 7.8k resistance when i have the black probe on the side with the line and infinite when i reverse the probes. so i guess its doing its job as a diode... is 7.8k resistance too much? how do you tell what resistance the diode should have in the direction it is allowed to flow?

[triplefour]

Re: Dell 1800FP no 5v standby.

R924 which is violet, green, black (7, 5, x1) should be 75 ohms and is 75 ohms so i see no reason to take it out for testing.

R928 (red black black) should be 20 ohms and is 20 ohms

R925 (Brown black orange) should be 10kohms and is 9.87k tolerance is 5% so this is within limits.

R903 (orange white orange) should be 39k and is 37.3, just above tolerance

R923 (orange orange silver gold black) i have no idea how to decode since silver is not supposed to be used as a multiplier. but i read only .6 ohms... and this is the fat resistor that goes right to the negative leg of the main filter cap so i was kind of expecting some high resistance there. i guess ill have to desolder one leg and check it out.
ok did that... same reading ... i have no way of knowing if this is right since i cant understand the color bands!

ok thats all the resistors and diodes in the network i drew out. hope i didnt miss any!

[Agent24]

Re: Dell 1800FP no 5v standby.

with the diode out i get 7.8k resistance when i have the black probe on the side with the line and infinite when i reverse the probes. so i guess its doing its job as a diode... is 7.8k resistance too much? how do you tell what resistance the diode should have in the direction it is allowed to flow?

Diodes aren't measured by resistance, they are measured by voltage drop. Use the diode test function on your meter instead

[triplefour]

Re: Dell 1800FP no 5v standby.

ok in diode mode i get 660 one way and inf the other way... what does this 660 mean in terms of voltage drop?

[Agent24]

Re: Dell 1800FP no 5v standby.

The diode test shows voltage drop across the tested component in terms of milivolts. So 660 = 660mV = 0.66 volts which is typical for a standard silicon diode.

Inifity in the other direction is OK. The diode is probably fine.

[budm]

Re: Dell 1800FP no 5v standby.

"with the mosfet removed i am now testing the diodes and resistors that were connected to the gate pin and i find that for d907 i get 95.2ohms resistance going either way. does this mean the diode has failed? or is it still "in circuit" ? only one way to find out i guess." If you look at the circuit you draw, that Diode d907 is in series with R928, this series connection is connected in parallel with R924, so that is why you are reading in both direction since those resistors are low value resistance. If you build that circuit on the bread board and make the measurement then you will understand why.

[triplefour]

Re: Dell 1800FP no 5v standby.

yes i have some homework to do but im starting to get it thanks to this forum! so um.. how do i be sure that the mosfet has failed?
and what about R923 with such a low reading (.6) ?

[triplefour]

Re: Dell 1800FP no 5v standby.

i understand now, because it is testing the resistance not through the diode but around the diode through the other connections

[budm]

Re: Dell 1800FP no 5v standby.

"and what about R923 with such a low reading (.6)" That is the very low value (<1 Ohm) for the Source pin of the MOSFET, the voltage developed across that resistor when the MOSFET is turn on is then used to feed the Isense pin4 of the ICE to monitor the current draw, if too much current flow through the resistor, the more voltage will be across the resistor, when the voltage goes beyond the set point, the IC will go into shutdown.

From spec sheet of the IC:
Isense (Current Sense)
The Current Sense pin senses the voltage developed
on the series resistor inserted in the source of the
external Power Switch. When Isense reaches the
internal threshold of the Current Limit Comparator, the
Driver output is disabled. By this mean the Over
Current Detection is realized.
Furthermore the current information is provided for the
PWM-Comparator to realize the Current Mode.

[triplefour]

Re: Dell 1800FP no 5v standby.

so then that resistor is fine? how can i figure out the color code though? orange orange silver gold black

[triplefour]

Re: Dell 1800FP no 5v standby.

also i still would really like to know why my resistance measurements change unpredictably when i change the scale?

[triplefour]

Re: Dell 1800FP no 5v standby.

ok using my ESR meter (which tests other things too...mosfets included...i hope) this is the reading i got.

c= 3.59nf
vt= 4.76

it does identify the Source Drain and Gate pins correctly

and seems to call it a N-E MOS
what is an N-E MOS?

and what does it all mean? would i need some info from the datasheet to compare against?

[budm]

Re: Dell 1800FP no 5v standby.

"so then that resistor is fine? how can i figure out the color code though? orange orange silver gold black" 0.33 Ohms, just remember that your meter test leads also have resistance so you you have to subtract the test leads resistance from what you read.

[budm]

Re: Dell 1800FP no 5v standby.

It telling you that it is N-Channel MOSFET.

[triplefour]

Re: Dell 1800FP no 5v standby.

ok i see how the color code makes .33 and how my meters internal resistance of .3 would add up to make .6 so i am sure that resistor is fine. but what of the readings c=3.59nf is this capacitance?
and what does vt stand for?
basically is this mosfet good or bad, or can i even know from these readings?