Maker Pro
Maker Pro

Output voltage is higher than it should be

pharaon

Oct 28, 2014
439
Joined
Oct 28, 2014
Messages
439
it's the power board for satellite receiver
the out put is higher than it should be like for the 23V it out 37V and the 5V out is 7V and the 12V is 18V
so which part is responsible for this high voltage output
i replaced the big brown capacitor which is 560V before the one beside the 8 legs IC because the old one got damage





 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
Looks like this is a switch mode power supply. These not seldom require a minimum load current to operate correctly. Did you measure the voltages with or without load?
If without load, connect some load ´, e.g. resistors, to the output to have some nominal current flowing. Then measure the voltage again. Chances are that everything is in good order.
 

Kiwi

Jan 28, 2013
471
Joined
Jan 28, 2013
Messages
471
I am speaking from experience here, so please don't take offense.

Is your voltmeter giving erroneous high readings due to it having a flat battery?
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
The regulating IC has a reference (feedback) voltage on pin 4 and this will be derived from ONE of the stabilised outputs. You will have to trace the circuitry back from pin4 to establish which supply line is being used as the feedback signal. It feeds the opto-coupler.

Check ALL the supply lines - the line being monitored will be the most accurate i.e stable output (I seem to see a 3.3 output????) and it may be overloaded.
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
Board looks dark around the switching ic.
The IC contains the main switching device (MOSFET) too and given the fault condition seems to be overloaded hence the darkening (I reckon).

May be useful to change all the electrolytic capacitors too.
 

pharaon

Oct 28, 2014
439
Joined
Oct 28, 2014
Messages
439
The IC contains the main switching device (MOSFET) too and given the fault condition seems to be overloaded hence the darkening (I reckon).

May be useful to change all the electrolytic capacitors too.
so do i need to change the IC and the rest of the capacitors ?
 

73's de Edd

Aug 21, 2015
3,613
Joined
Aug 21, 2015
Messages
3,613
Sir pharaon . . . . .

This is what I suspicion . . . . . . over an extended time period of use . . . has happened with that unit of yours . . . .
A RUN DOWN OF ITS OPERATION . . . .

Referencing to the supplied FAIRCHILD tech application note, to then have its reference schematic.
Observing far left bottom corner and rising upwards, you see the origin of the MAIN AC to DC power supply of the unit and it entering into the C102 supply capacitor being across BLUE A and BLUE B. The DC voltage is now being upwards of 300VDC, then routes to the right from BLUE A over to BLUE C. There is a power flow path from BLUE C to BLUE D of the primary of the T101 switch mode power supply transformer . Then there is the BLUE D connection made to the left to the BLUE E connection to the drain connection of an internal POWER MOSFET *** built within the IC proper.

*** ( THIS POWER FET can handle upwards of 22 WATTS, so you can see the long time discoloration effect upon the nearby phenolic PCB area around it.)

All is in order for that power supply to spring into operation . . . . being only less the drive pulses into the FET.
BUT . . . that voltage would have already been initiated, at the instant that the unit was plugged into AC line power.
The voltage at BLUE A would have passed thru R105 dropping resistor and fed into the IC pin 5 where there is an internal zener to feed initial internal supply voltage to get a timed burst of drive pulses to the gate of the power FET.
In this initializing start up procedure, the current consumption of the BLUE C-D primary winding is sampled, as well as current sensing at I.C. #4 and voltage sensing at I.C. # 3.
Simultaneously the power drain on the transformer by the secondary supplies created by D203-D204-D207 and D205 have to be within established limits.
If all are in order, then the dedicated supply voltage source for IC101 will be active.
That consists of the BLUE F-G winding and the D102 rectifier and the filter capacitor C106.
( If a sluggish turn on procedure has time evolved . . . expect a capacitance depleted C106. )
If this system now is constantly staying on and running after start up, then we need to inspect your across the board error in regulation.
That concerned area will be within the PURPLE BOX area.
It interfaces thru 5 interconnects to associated circuitry.
There is the far left 817 optical isolator coupler, with its photo transistor that can progressively start pulling the Vfb voltage at IC pin #3 to the HOT ground.
Less voltage at Vfb tells the pulse width of the drive to the FET to shift as a voltage compensation / correction.
MEANWHILE at the other right half of the optical isolator, with its internal LED its normal brightness is established by its BLUE I supply voltage and the conduction of the IC301 precision adjustable zener.
You can see a GREEN circled I prime supply point that flows down to BLUE J and encounters its need to then flow thru two divider resistors of a voltage divider bridge consisting of R203 and R205.
That derived voltage fed into IC301, then establishes the conduction /brightness of the internal LED that is fed by BLUE I .
The constant monitoring and interactions in compensatory shifting the frequency and pulse widths of the drive to the power FET give you a constant voltage from its different supplies and a response to shifts of loads on the supplies, within its design limits .
How you interpret the above info, relates to the fact that the most important section of stabilization relates to the PURPLE BOX area.

And where is it obtaining its all important power ? . . . . . . The GREEN CIRCLE areas of the 5VDC supply area.

If those HARD worked filters . . . . .having had the HELL hammered out of them by incoming 10's to 100's of thousand pulses PER SECOND. . . . have not succumbed and are starting progressively dropping capacitance by incremental drying out of their internal electrolyte.
Then, fully expect a gradual on setting of progressive heating to further speed up the capacitance decay process.
Of the suspect C213 and C214 electrolytics of that 5VDC supply, the first C213 is most subject to the hardest HAMMERING, while the C214 receives some pulse buffering by the inline 4.7 ufd inductor.

So, should you choose to test this unit . . . the approved Mac Guyver technique . . . will need you to do this.

Choose your favorite and most used finger upon one hand . . . . . then revert to actually USING the like finger on the OTHER HAND.

Fire up ye olde power supply . . . . . test the temperature of the IC with aforementioned finger until you can DEFINITELY detect a warming of the units top case.
Hopefully after this warm up time you can then move over to the COLD secondary side of the power supply and do a fingertip temperature test to the top metal caps of each of the 9 caps on that side of the transformer.
At the end, go to the 1 electroltytic on the HOT side of the board near the IC and check its temperature. ( a 47 ufd 50VDC ? unit)

If you find hot ones, solder a like value of electrolytic capacitor across its connections and do a voltage check of your supplies again.

If this gets you fixed . . . . . a MASS change out of all of the electrolytic capacitors would be in order.

CIRCUIT APPROXIMATION OF WHAT YOU HAVE THERE . . . .

upload_2018-1-7_3-53-27.png

73's de Edd
 
Last edited by a moderator:

pharaon

Oct 28, 2014
439
Joined
Oct 28, 2014
Messages
439
test the temperature of the IC with aforementioned finger until you can DEFINITELY detect a warming of the units top case
it was just little bit warm

test to the top metal caps of each of the 9 caps on that side of the transformer.
only one was too hot that i can't keep my finger on top of it and the rest were cold not even cold

At the end, go to the 1 electroltytic on the HOT side of the board near the IC and check its temperature. ( a 47 ufd 50VDC ? unit)
it was cold

 

73's de Edd

Aug 21, 2015
3,613
Joined
Aug 21, 2015
Messages
3,613
Sir pharaon . . . . .


NOW, look closely at the top of that cap . . .wouldn't suspicion anything at all . . . would you ?
People tend to discount that internally heated and thus vaporized electrolyte can "blow by" wire leads and the seal of a loosened neoprene plug used at the bottom of the of the electrolytic case.

Since your " Lab Equipment Shelf " is being so very-very sparse, lets take this situation for a further evaluation of your DVMs test capability . . . . specifically its AC frequency range limitations.

We would be wanting to have the power supply in its same "BAD" condition and your placing the DVM in AC measuring mode and in maybe the ~ 10VAC range . . .unless it is Auto ranging automatically for you..
Then you power the unit up and measure the AC voltage being read across that now capacitively deficient RED circled unit.
Log it down and then note the right sides nearby GREEN 10 uh inductor that couples power over to the the like sized electrolytic. Take the voltage reading across that second companion capacitor and compare the two sets readings and typically expect a much larger reading at the first tested capacitor.
Next take a new or known good capacitor of the voltage rating and capacitance of the second capacitor that you tested the voltage of.
POWER down the unit.
The old capacitor is still in circuit but do a temporary solder connection of the PROPER polarity connection of the new cap to the old caps connections.
POWER up the unit
Take the same AC reading across the second capacitors . . . . now tandem circuitry . . . log down and compare the readings with a GOOD cap additionally being in circuit.
POWER down the unit.
If the capacitor values happen to be the same for both old capacitors move that now unsoldered and pulled NEW ? GOOD capacitor over to do the exactly same two subjective tests and do the same voltage logging, but this time expect significant differences after the good caps connection into circuitry.

A final test would be to know exactly HOW . . . YOUR meter . . . . responds with there being a DC voltage superimposed upon an AC voltage testing that you are doing.
In this case . . . assuming that the 5 VDC supply portion is involved . . . there is both the 5VDC supply voltage present, simultaneously with that measured AC voltage (ripple) from the power supply.

EXCESSIVE ripple BAD . . .VERY BAD. . . . as per Babu . . . .a . .la . . .Sienfield.

On some DVM's . . . erroneous AC voltage readings are received under those conditions of AC with a combined DC presence.

To test for this, insert a series capacitor between the meter lead reading and the circuit being tested.
The capacitor STOPS the DC from passing but a 0.1 ufd capacitor only has the attenuation effect of a 15 ohm resistor at 100KHZ . . .or even better, a 1 ufd series cap acts as a mere 1.5 ohm attenuation of the AC signal.
Whichever one you have available, but then go back and repeat the testing to see if the AC readings vary.
With your newfound familiarity of your DVM's AC voltage reading behavior, you may then use it to advantage in evaluating switch mode power supplies filter capacitors in the future.
In this one you lucked out in being able to use individual temperature evaluation of the different capacitors on the COLD supply side of the supply.

Thasssssit . . .

73's de Edd
 

pharaon

Oct 28, 2014
439
Joined
Oct 28, 2014
Messages
439
sir 73's de Edd

i did change the heated capacitor and test the receiver and it work find
the next step i did were changing the rest 8 capacitors
and when i try the device again it's not working anymore and the output voltage now is too low
is i change back the 8 capacitors to the old one but kept on the new 10v 1000mf new capacitor that the device work with before but still the same issue
the device is not working and the output voltage is too low like the 23v is 3 v
and the 3v is almost 0v

so what do you suggest that i do
 

73's de Edd

Aug 21, 2015
3,613
Joined
Aug 21, 2015
Messages
3,613
Be sure that the polarity is not reversed on one or more of your capacitor installs, or having a bad capacitor if used capacitors are being used .
Also do a power off test of all of your diodes, particularly the one associated with rectifying the 3.3 VDC supply.
 

pharaon

Oct 28, 2014
439
Joined
Oct 28, 2014
Messages
439
ok it was faulty diode the one beside the red circled old faulty capacitor and i change it. now it's working...but will check for the output voltage and see
 

kellys_eye

Jun 25, 2010
6,514
Joined
Jun 25, 2010
Messages
6,514
ok it was faulty diode the one beside the red circled old faulty capacitor and i change it. now it's working...but will check for the output voltage and see
Note that they are high speed rectifiers and you cannot use an ordinary silicon diode.
 

73's de Edd

Aug 21, 2015
3,613
Joined
Aug 21, 2015
Messages
3,613
Sir pharaon . . . .

Looks like a bit of familiarization with components used on your board are now in order.
In looking at your last photo I am seeing the derivation of 5 VDC---3.3VDC---12VDC and 23VDC from this power supply down at the bottom WHITE connector.

ELECTROLYTIC CAPACITORS . . .

Without a bottom foil path shot of the board and the voltage rating and capacitances of all of the electrolytic's , I now can ONLY give a well Edd-jum-i-cated guess of the top of the boards flow paths.
The boards heaviest demand seems to be the RED circled "BAD" capacitor and then an interconnect thru a 10uh filter inductor to the other companion E- capacitor down the RED line.
Assumption is, that the DC input to one of them is from a RED circled " C" diode or both, if being used in a full wave rectifier configuration.
The 105 deg C temperature rating is correct on the right capacitor, one should be expecting the same temp rating on the left one ?.

Moving to the GREEN line supply below , the same Cap-inductor-Cap power flow but I cannot associate with their supply rectifier diode.
They are using an 85 deg C temperature rating vice a 105. Only the 25VDC rating viewable . . .BUT . . . ole' Sam Young certainly lets his presence be known, on his products markings.

Moving to the BLUE line supply, far left, considering that the very nearby B diode is being its rectifier, it then enters into a 105 deg C input capacitor, and if the cap past the 10uh inductor has never been changed out , it is being a 85 deg C .

??????? Wrong capacitor installed during a later repair of the unit or did the manufacturer gamble that he could cut corners on that E-caps rating due to the HF buffering provided by the intermediately located 10 uh inductor. ?????
I would be using 105 deg C caps at both positions.

Same basic writeup / observations made on the YELLOW line supply and the BROWN line writeup, except no intermediate inline filter inductor is being involved with their circuitry.

DIODES . . .

On switching power supplies you need to use special fast recovery diodes.
Using a standard diode will result in the diode not stopping its one way flow path in FAST enough time and the equivalent effect of applying AC to its supply E-filter . . . that's a NO NO . . . .for the overload factor then imparted to both the diode and the subjected Electrolytic.

On this units rectifier diodes I see three different current capabilities being used. There are the minimal current units being used in the two A positions and then a medium current diode as the B diode and a final much higher current being used for the two C diodes positions.

For the A diodes the UF4007 product number of diode would suffice, whereas, as for the B and C diodes you need to pass us the numbers for researching.

I take your faulty diode as being the first large C diode instead of the smaller A diode just to the LEFT of the bad cap.

A bottom PCB foil path photo along with all diodes marked on I.D.numbers and an additional. one lead out of circuit and specifying the Vf being read on your DVM in its special diode test position would give us a bit of a clue of the 4 diodes types.
  • Common diodes read in the 600-700 millivolts of Vf.
  • Fast switching or recovery diodes tend to fall down in the 400-500 millivolts range of Vf.
  • And the final set of large diodes may want to exhibit FAST switching along with high current capability and low voltage drop, that might incorporate a Schottky family of diode with its low 200 millivolts of Vf.


A FURTHER PHOTO MARK-UP . . . .

POWER_SUPPLY_E-_CAPS_AND_DIODES.png


73's de Edd
 
Last edited by a moderator:
Top