I have an adjustable power supply, It works well.
The voltage and current readouts use LCD displays, And the backlight has the color of a green LED, but has the heat of an incandescent.
I hope it's just the controls, or some other component.

The question is: Is it LED or Incandescent? I don't want have to replace bulbs in this.
Any answer would be great! 

My brother and my mother, had a Cason alarm clock, with incandescent backlit and cyan filter, which gave the illution of cyan LED. I think that it is your case here (Incandescent lamp + green filter).
These miniature incandescent lamps, usually have a very long life, as they are often underdriven.

Would a photo help?
And does the incandescent style have any uneven areas or is the intensity uniform?
We also have an alarm clock that has incandescents, But you can see 2 bright spots of light, And it's in a travel alarm clock with a single AA battery.

The backlight on the power supply has more of a "Retro" color, But I don't know for sure. The light is completely uniform in intensity.

Post a photo...

For the backlight source: I wouldn't care what is there (except of just curiosity), as long as it does it's job.
Now the display works, so no reason to bother about it. Once it eventually stops working, only then will come the time to really look inside.
Because what would you do, if you find either variant: Would you immediately upgrade the backlight to LED's? Or if the backlight is of very inefficient LED setup (some old type), would you upgrade it just now? I don't think any of these making any sense, if the displays do work well...

Don't forget, when the power supply is from the 90's or older (I mean it's design, not the real manufacture), the incandescents are very likely light source - except of red, the the LED's were definitely less efficient at that time, so small incandescents were really the "workhorse" of such instrument illumination.

I purchased a Samsung blu ray player yesterday which also has Netflix and other useful features to me. I noticed the display has a cool white glow to it and think it may be LED as it doesn't look VFD.

See here.

Post a photo...

For the backlight source: I wouldn't care what is there (except of just curiosity), as long as it does it's job.
Now the display works, so no reason to bother about it. Once it eventually stops working, only then will come the time to really look inside.
Because what would you do, if you find either variant: Would you immediately upgrade the backlight to LED's? Or if the backlight is of very inefficient LED setup (some old type), would you upgrade it just now? I don't think any of these making any sense, if the displays do work well...

Don't forget, when the power supply is from the 90's or older (I mean it's design, not the real manufacture), the incandescents are very likely light source - except of red, the the LED's were definitely less efficient at that time, so small incandescents were really the "workhorse" of such instrument illumination.

Sorry that I took so long.
Taking photos happens to take a really long time, And the colors look different than what it really looks like. Then I remembered the website for it, And that likely looks better.

The light is uniform, And a light green color. I think it looks more like an LED color, But that can't explain the heat coming from it.
The design looks to be recent, And it has replaceable componets.
Here is the "Data Sheet": http://www.extech.com/instruments/resources/datasheets/382200_382202data.pdf

The website for it is no help, All there is is "Backlit LCD screen".
If the backlight happens to be really old LEDs, I will keep it in there.

Some LCD's have a greenish filter layer on the back in order to vgive them the classic "LCD look". Maybe some of those filters really make the light look like LEDs

Never seen those filters before. Could the heat be caused by the power needed to drive the LCDs?

Edit: I just did the CD spectrometer test, And I saw a little deep green, Lots of light green, And a little bit of red. That indicates that the backlight might be more likely Incandescent than LED.
Maybe I coul contact the company and ask them what the backlight really is.

Edit again: I peered in, And saw that it uses the ICL7106 chip.
It can directly drive LCD or LED displays. It is being used for the LCD function.

Edit once again: I peered in again, And I saw that it uses l7809cv voltage regulators. Because it is linear, It creates a lot of heat, And that could explain the heat from the displays.
I also saw that the displays are very slim, And have a typical design of LED backlit displays. Or the heat could be from the LCD drivers.

Well, the 7106 can drive only LCD, it has not enough drive strength to drive the LED's, neither with segment outputs, nor the internal supply scheme.
The 7107 is the 7106's brother designed to drive common anode LED's. The difference is not only in the stronger and only low side segment drivers, but also in the supply scheme:
The 7106 has the ~5V (below B+) for the digital part and display driver generated internally, the 7107 needs an external supply (the positive part vs GND).
The 7106 has on pin 21 the LCD backplane driver, for 7107 that pin serves as the negative connection of the supply for the digital and display (it is named GND).

For now, I will assume that it uses LED backlighting.
Edit: I know how to know if it is LED or not.
Measure the afterglow when I turn the power supply off.
Incandescent should trail off, And LED would virtually be instantly off.
I'll have to do that.

Update: I did the test, And when the power supply was turned on, The backlight came on like a capacitor was charging, And when I turned it off, It went down in a very linear way. Since Incandescents trail off in a non linear way, This is edvidence that the backlight is LED.

Update: I did the test, And when the power supply was turned on, The backlight came on like a capacitor was charging, And when I turned it off, It went down in a very linear way. Since Incandescents trail off in a non linear way, This is edvidence that the backlight is LED.

That is very stupid design, it means the ICL7106 are exposed to that higher temperature as well.
One big disadvantage of the 7107 - the power dissipation from driving the high LED currents heats up the silicon, what means the reference of the AD converter drifting. And that means the meter accuracy gets worse.
The low LCD currents are supposed to prevent any heat buildup, so make an expectation of the 7106 to be more accurate. But when someone design the equipment so the chip is heated externally, the accuracy is gone as well.
You may say the thing may be adjusted after warmed up, but once you open the case to do that, the temperature profile is different from what it is there when normally running, so the setting is not much more accurate.
They may use a separated reference IC, but the difference towards the internal reference is not that big (it helps to fix the 7107 DNL error caused by different power dissipation for different displayed numbers by not letting the related heat to reach the reference).
The only reliable way is to use really low power dissipation around the accurate stuff and that is not fulfilled there.

I guess that explains the +-1.5% to +-2.0% accuracy.
You would have to use a Multimeter for Low-Voltage-Tolerance applications.

If it was at 24 volts, Then the display could be off by .48 Volts!

I guess that explains the +-1.5% to +-2.0% accuracy.

Well, that is typical accuracy grade for the lower end hand held meters, so it won't improve the accuracy that much.

With that hot background I would expect worse, somewhere in the +/-5% range.
But it is true, they may use a separate dedicated voltage reference IC (instead of the Zener based one build in the 7106).
The dedicated reference IC's are usually band-gap based designs, so rather easy to factory-trim for flat temperature profile (the physics makes all process variations canceling out by just a single temperature point factory trimming for a given target voltage; the target selection then steers the resulting temperature profile, but still it remains independent on the silicon fabrication spread; with Zeners the varying parameters have different temperature coefficient, so you would need at least two temperature trimming for each individual component; that is just totally not feasible in the mass production, unless we are talking about a component selling for $10 or more for 1mil/year ordering quantity, so with voltage references it would be way better than 0.05% complete inaccuracy, include temperature and aging).

Quote from: wattMaster on May 01, 2016, 11:02:01 AM

I guess that explains the +-1.5% to +-2.0% accuracy.

Well, that is typical accuracy grade for the lower end hand held meters, so it won't improve the accuracy that much.

With that hot background I would expect worse, somewhere in the +/-5% range.
But it is true, they may use a separate dedicated voltage reference IC (instead of the Zener based one build in the 7106).
The dedicated reference IC's are usually band-gap based designs, so rather easy to factory-trim for flat temperature profile (the physics makes all process variations canceling out by just a single temperature point factory trimming for a given target voltage; the target selection then steers the resulting temperature profile, but still it remains independent on the silicon fabrication spread; with Zeners the varying parameters have different temperature coefficient, so you would need at least two temperature trimming for each individual component; that is just totally not feasible in the mass production, unless we are talking about a component selling for $10 or more for 1mil/year ordering quantity, so with voltage references it would be way better than 0.05% complete inaccuracy, include temperature and aging).

My multimeter has .8% accuracy.
If only they could add a fan to it...