Maker Pro
Maker Pro

Variable AC Isolated Transformer (showing wrong voltage)

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
I went ahead and adjusted it back to where I started which was 120V = 142V. This schematic may be helpful at finding proper leads. I will be able to test this in a few minutes. I am assuming because of d3 d4 d5 d6 it is a rectifier. So I will be testing an AC signal on the input pins?

schematic.jpg
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I went ahead and adjusted it back to where I started which was 120V = 142V.

That's good. Wrong, but back to a known position. :)

This schematic may be helpful at finding proper leads. I will be able to test this in a few minutes. I am assuming because of d3 d4 d5 d6 it is a rectifier. So I will be testing an AC signal on the input pins?

I think the AC IN is a low voltage AC from which the supply rails fo rthe meter itself are derived.

The actual input appears to be on the top right corner, and these are wired to one of the connectors on the board.

My impression is that this meter is reading DC.

That would explain the rest of the board as a precision rectifier (although I'm not sure you need to go to all that effort).

Did you mention if you had tried this unit with an output load?

It would be interesting to get 2 low wattage lamps, connect one to the mains and one to the output of this and adjust the variac until both have the same brightness -- then measure the voltage and note what the built-in meter says.

The huge transformer inside this unit is way too large to just power the electronics, so I wonder if it is not also an isolating transformer for the variac. Having these 2 inductors connected together could cause unusual things to happen to the AC waveform under no-load conditions.

I'd suggest you look at the output waveform (I've thought of asking this before) but:

1) we need to know if the output is isolated
2) you need to have appropriate probes for one of your scopes
3) you *REALLY* need to know what you're doing.

A mistake in any one of these three can cause a large bang and the loss of your scope (or worse)
 

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
Did you mention if you had tried this unit with an output load?

It would be interesting to get 2 low wattage lamps, connect one to the mains and one to the output of this and adjust the variac until both have the same brightness -- then measure the voltage and note what the built-in meter says.

This is an absolutely brilliant idea! I will do this and let you know the results.

The huge transformer inside this unit is way too large to just power the electronics, so I wonder if it is not also an isolating transformer for the variac. Having these 2 inductors connected together could cause unusual things to happen to the AC waveform under no-load conditions.

I'd suggest you look at the output waveform (I've thought of asking this before) but:

1) we need to know if the output is isolated
2) you need to have appropriate probes for one of your scopes
3) you *REALLY* need to know what you're doing.


1) Yes the output is open ground. I even tested for this.

2) They are TP6100 probes. Rated at 300V for X1 and 600V for X10. It seems the graph shows you are in the clear for less than 5 MHz on X1 and in the clear for less than 1.2Mhz for 600V.

3) I really appreciate your concern. I really want to learn this stuff and I want to be safe. I am taking precautions by knowing what I am probing (such as going on the net writing on the forum and learning what things are first). I am using one hand, keeping my area clean. The truth is I cannot stop doing this. I hate to say I will take my chances... but thats what I am going to do. I will take all tips and pointers and what I already know and be as safe as possible.

Just so you know I am extremely interested in being safe. I want to read or watch a video on it or a series of it. If you know of anything let me know. I will take the time to learn more.
 

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
#3) Really what I want to do is fix and tweak this stuff the best I can. Sell it and start on the breadboard with low voltage ac and dc currents and voltages. For now I want to get this variac up to spec if possible and sell it. I need to make back some money I have too many units sitting here. I do not want to give the stuff away if I do not have too.


So if my last post sounded pompous and looked as if I like being foolish. This was not my intent. I just want to finish these projects. And I DO appreciate everything that has come from these posts!
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
OK, presumably you've checked that each output is isolated from ground, live, and neutral.

If this is the case, you can connect your oscilloscope across the output.

However, here's a few precautions:

1) use the x10 probe
2) set the output to a low voltage (say 20V as read by the fluke)
3) ensure the scope is set to the highest v/div.

Then reduce the v/div until your display takes up about half the screen height.

Does it look like a nice sine wave? Can you measure the peak to peak voltage?

What is it, what does the Fluke read, and what does the meter on the variac show?

Can you provide a picture of the trace?

You probably need to know how to operate the scope too. That's beyond my abilities. Maybe it has a magic "setup" button that picks a good set of parameters (mine does, and it usually works).

Then see if adding a load makes any difference
 

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
OK, presumably you've checked that each output is isolated from ground, live, and neutral.

Yes. NO continuity from mains to variac output for ground hot and neutral.


If this is the case, you can connect your oscilloscope across the output.

However, here's a few precautions:

1) use the x10 probe
2) set the output to a low voltage (say 20V as read by the fluke)
3) ensure the scope is set to the highest v/div.

Then reduce the v/div until your display takes up about half the screen height.

I hooked the scope to the hot wire and the scope showed a signal. I did not attach ground at all. Why would something show up without ground? As far as I know the scope ground is definitely attached to the earth ground, but the variac output is not attached to earth ground. So why was there a signal at all?

Here is what I got..... (i turned off all display info and turned on the grid. I also adjust the variable div/sec so it showed one complete sinusoidal)





Now I wanted to see if it changed when hooking to mains hot. (I also did not hook up ground because I figured it would be redundant.) Notice the wave is bigger.



Does it look like a nice sine wave? Can you measure the peak to peak voltage?

The variac output showed 17.5V P-P. Since this is 10x I am unsure what is going on here.

The mains output showed 35V P-P. I am confused by this since the DMM said the same


What is it, what does the Fluke read, and what does the meter on the variac show?

Fluke was at 120V no load. Variac at 128V no load.
With a load the Fluke was out 118.5 and the Variac was 127.

So it did move a volt or 2 with a load. Weird.



Just so you know. I am pretty good with an oscilloscope and I understand all the basics. I even understand AC and DC coupling and why to use them. Apparently I did not master the intensity knob on the Tek 2235!!! haha
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
You do need to connect the ground, otherwise the voltage you're reading is due to coupling to ground.

You need to be absolutely sure the outputs are floating though, because the ground lead of your oscilloscope is really grounded.

If the scope is set for 0.5V/division, the voltage in the last image would be about 3.5V, which you then multiply by the probe -- in this case 10x - giving you 35 volts peak to peak. That is about 12.3V RMS (divide peak to peak by 2 x sqrt(2) if the signal is a sine wave).

The signal is slightly distorted, but it doesn't look severe.

When you load the output, does the multimeter *and* the variac read a lower voltage?
 

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
You do need to connect the ground, otherwise the voltage you're reading is due to coupling to ground.

Does coupling to ground mean through the inductance of the isolation transformer in this particular case?

When you load the output, does the multimeter *and* the variac read a lower voltage?

Yes the variac display moved by a volt or so and the fluke moved like 2 volts when a LOAD was added. What does this tell you?


Pictures worth 1000 words....


WITHOUT LOAD




WITH LOAD (notice this is this little ripple... perhaps this is just the device I plugged in?)

 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Does coupling to ground mean through the inductance of the isolation transformer in this particular case?

Well, stray capacitance, but you get the idea.

Yes the variac display moved by a volt or so and the fluke moved like 2 volts when a LOAD was added. What does this tell you?

It tells me that the problem is probably not because it is unloaded.

The displays on the CRO tell me that the waveform is not distorted and that doesn't explain the difference in reading.

So, we've eliminated some things :)

Let's go back to measuring the voltage on the input of that display and comparing it to what the display is showing.

You should be able to use the CRO to determine if the input is AC or DC too.

This step is to determine if the display is reading the wrong voltage, or the voltage being applied to it is wrong.
 

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
Let's go back to measuring the voltage on the input of that display and comparing it to what the display is showing.

I like this idea. I am having trouble understanding how to probe this though. Can you help me understand the pins. I see

on the first set of pins is 1 2 and 3 and it has D.P written by it. I imagine Digital Panel.
The other 1 - 7 has : ~ ~ gnd low hi LED LED

What is Low and Hi?
What is the two ~ ~ pins? just ac signal.

And the schematic ->>> HERE

Can you help me understand this?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
low and high are the inputs I believe.

dp is decimal point.
 

AcousticBruce

Apr 1, 2013
43
Joined
Apr 1, 2013
Messages
43
I honestly do not know what has happened... unless testing and probing the meter circuit reset something... but now it is working fine.

I am going to sell this now... consider this case closed. Thank you for all the help.


Keep a look out for all my transistor, capacitor, resister and opamp questions that are going to soon follow. :p
 
Top