RFG70N60 Mosfet testing...

[Landon]

I am trying to test two RFG70N60 Mosfets, with my meter set on testing
diodes one Mosfet comes up with a reading of 0.001 the other 0.003, this
seems kind of low.

These are both out of a UPS that is having problems, does anyone know
what the readings for the Mosfets should be? I emailed Fairchild
Semiconductors but they never replied.


Thanks,

-Landon

 

[John Musselman]

Landon,

Put your meter on ohms and measure between the drain and the source. These
are the righthand two leads. You should get a high reading, either no
reading or high megohms. If not, it's shorted, the usual failure mode.

John Musselman

 

[Landon]

Hi, thanks for replying.

I found a procedure online somewhere that basically said the following:

------------------------------------------------------------------------
-Set meter to check diodes.
-Put positive lead on the GATE, put negative lead on the SOURCE (this
will charge the gate, before doing this the DRAIN and SOURCE pins should
give a HIGH reading).
-Then move the positve lead to the DRAIN, it should give a LOW reading.
-Touch the GATE and SOURCE pins with finger tip, the gate should
discharge, then the DRAIN and SOURCE pins should again give a HIGH reading.
------------------------------------------------------------------------

This is where I am stuck, the Mosfets are acting the way the test says
that they should but I am used to seeing diodes reading at 0.5xx to
0.7xx, these are reading at 0.001 and 0.003, I'm not sure if this is too
low or right where it should be.

-Landon

 

[John Musselman]

Landon,

That's a cute procedure and may even work.

In the "diode test" position, your DVM is generating a small current and
reading what voltage is generated. So a diode will read out the voltage
across the junction (the "forward voltage" or Vf) with this current passing
through. In a diode, this tends to be 0.5 to 0.7 for a silicon diode, and
varies little with the current.

In a MOSFET, there isn't a diode junction. It's just like a resistor, so a
low reading is a low resistance. (A low voltage is generated by the test
current). The low number is OK, if you have, indeed, charged the gate. You
should get a high reading (or overload, or infinity) when the gate is
discharged or shorted to ground.

John

 

[Landon]

Thank you for the reply...

I believe that answers my question (that the 0.001 to 0.003 readings
are normal), and yes when the gate is discharged it is giving an
infinity reading.

I guess I will keep troubleshooting this UPS, my next guess is that the
LM324 that is driving the two MOSFETS isn't giving them the correct
waveform or isn't being told to give them the correct waveform.


Thanks again,

-Landon

 

[loedown]

What symptom is the unit showing?

Paul

 

[Landon]

Everything works fine when it is plugged in, it stays at 120VAC, when it
looses power during a power outage (or if you just simply unplug it) it
immedietly plummets to 106VAC and then keeps dropping until the battery
goes dead.

After plugging it back for a while the battery recharges and stays
around 12.8VDC, even after being unplugged and removed from the case the
battery still stays charged at just over 12 volts. The windings in the
transformer appear to be fine (visually and when ohm tested), nothing is
obviously burnt or smells burnt.

The windings for the transformer have 12 volts going to the center tap
and then each size of the center tap is turned off/on with the two
mosfets to simulate AC voltage, they are in turn driven by an LM324N,
both gates of the Mosfets have about 2.5 volts on them but I have no
idea what the wave form looks like.


-Landon

 

[loedown]

The SLA battery you are talking about is most likely going to be the
culprit, although it stays at 12V after you unplug it, you have to check the
battery under load conditions. Very common for the battery to die and the
explanation that it starts at 106VAC and goes down hill from there fairly
rapidly points squarely at the battery, UPS draw reasonable current and the
battery can't keep up. Mind you, I don't have the unit in front of me.

Paul

 

[John Musselman]

I agree with Paul. I think the battery needs to be replaced.

John

 

[Pig Bladder]

What symptom is the unit showing?

Absense of context.

 

[Landon]

Thanks for the reply, the battery going dead early would cause exactly
what it is doing, I been going down the path of the frequency wasn't
correct to make the MOSFETs switch on and off fast enough to create the
proper voltage coming out of the transformer.

A couple of things came to mind...

1. The 10:1 ratio in the transformer would create 106 volts if the
battery had suddenly dropped to 10.6 volts (normally this would be
120VAC from 12VDC), if the the battery voltage kept doing down so would
the output voltage.

2. When I equated this to the last time my car would not start because
the battery was trying to die it all made sense.

When I get the thing re-assembled I'll monitor what the battery voltage
is doing when the unit looses power.

More later...

 

[Landon]

Hi,

Well, I don't know what to think now.

The battery is holding at 12.4 volts with an output voltage of 102VAC
with no load on it at all.

I let it recharge for a while, the output was holding at 120VAC with no
load on it for over five minutes with the battery still above 12.5
volts. When I plugged my 140watt soldering gun into it and pulled the
trigger it promptly shut itself off.

140watts at 120VAC only comes to 1.17amps which should not cause the UPS
to shut down, when this UPS was still functioning normally it had a
computer, monitor, 60watt lamp, and power supplies for an answering
machine, computer speakers, and a PALM recharger all plugged into it and
would still run long enough to shut Windows down.

I waited a few minutes then turned it back on, the battery voltage is
still over 12 volts but the output has dropped back down to around 100
volts.

Any suggestions?

 

[loedown]

Measure the battery's voltage when the unit is on load, not when it is out
of circuit, not when it is charging or at any other time, this is the only
way you will see if it is truly the battery. Even if this isn't possible,
get a 12V filament lamp and connect that to the battery out of circuit and
monitor the voltage. Without doing these things, you are just compounding
the issue and not determining whether the battery is at fault, or the unit
itself.

SLA charging voltage should be around 13.8 V

Paul

 

[Landon]

Hi again,

I let the battery charge all last night and all day until just a few
minutes ago, it was charging at 13.8 volts. With a 25watt soldering
iron plugged into it the lowest I could get the battery to drop was 12.2
volts. I plugged my 100/140watt soldering gun into it and put it on
100watts, within about a second the voltage plummeted at 10.something
volts and then the unit shut itself down.

I let it charge again up to 13 volts, as soon as I put it on 140watts
the unit instantly shut down. On 100 watts the draw is slow enough to
notice that the voltage plummets to 10.6 volts in about 3 seconds, then
it shuts down.

This is almost the identical test that is done on car batteries, they
are supposed to stay above 10.5 volts for so many seconds, if they drop
below 10.5 too soon that battery is considered bad and in need of
replacement which is what I will be doing with this thing.

Thanks for your help, I never would have suspected the battery.


-Landon

 

[Rich Grise]

I let the battery charge all last night and all day until just a few
minutes ago, it was charging at 13.8 volts. With a 25watt soldering
iron plugged into it the lowest I could get the battery to drop was 12.2
volts. I plugged my 100/140watt soldering gun into it and put it on
100watts, within about a second the voltage plummeted at 10.something
volts and then the unit shut itself down.

That's it, then. It's the battery. You need a new one.

Sorry.
Rich

....

 

[Landon]

Hi,

I found the website for the company that makes the battery, the life
span for this battery "Vision CP1250" is 3-5 years, it lasted about 4 so
I'm not surprized that it is trying to die.

I found a local supplier that sells them for around $24.00(US).


-Landon

 

[Alan the Tech]

You might want to do a load test on your battery, using just a volt
meter will give you an incorrect assesment of the battery

To do the test you need to first pull the battery from your system.

Then there should be a label on the battery telling you the amp hour
rating of the battery. It's usally mark AH-12

With that info calculate the resistance needed to draw that current
from a good battery. volts/ current = resistance

Then calculate the power needed so you don't burn up the resistor
resistance * current squared = power and get a power resistor with
at least that rating.

If you have or can get a good high current relay or contactor put it
in series with the resistor across the battery.

With the volt meter across the battery close the relay and watch the
battery voltage for 30 seconds.

If it drops below 10 volts replace it.

Also I don't know what type of battery is in your system. The most
common ones are lead/ calcium commonly known as gel cell.

If they go below 60% or there abouts of capacity they will lose there
abillity to hold a full charge.

Most UPS and IPS system have an undervoltage shut down circuit when
the batterys go below a set voltage level.

This is the test I have used when I was working on IPS and UPS power
systems.