# Converting DC to a 60Hz AC signal

#### Management

Joined Sep 18, 2007
306
Essentially I have a meter that has 3 inputs. I am using it to measure voltage and current on a single phase 60Hz 208V line. The other two inputs are not being used.

The thing is a have a DC signal I would like to measure as well. That DC voltage can either be 3V, 6V. 9V, or 12V.

Basically I want to be able to use this meter to detect that DC voltage but it only accepts AC. Therefore is there a simple inverter (I mean simple) that I can build that will take that DC voltage and create a 60Hz sinewave of some voltage level for each DC voltage level? Meaning if the voltage level of the DC signal changes it would change the AC sinewave level and I can read it in a meter.

Reasoning: This meter has a TCP/IP connect and I don't want to add another device out there (better yet can't) so I want to use this device since I have remote capability.

Actually frequency should really matter. I only care about having a measurable voltage AC signal.

Thanks!

Last edited:

#### colinb

Joined Jun 15, 2011
351
Oh man, that is really a complicated way to measure your DC voltage. You may need to produce a true sine wave and make sure its rms voltage is the DC voltage you want to indicate. That means the peak-to-peak voltage is significantly higher. You'll need a non-trivial circuit to do this.

#### Management

Joined Sep 18, 2007
306
Oh man, that is really a complicated way to measure your DC voltage. You may need to produce a true sine wave and make sure its rms voltage is the DC voltage you want to indicate. That means the peak-to-peak voltage is significantly higher. You'll need a non-trivial circuit to do this.
First of all thank you for the quick reply.

To tell you the truth I don't need the RMS voltage of the generated AC signal to equal the DC voltage I want to measure. As long as I know what x amount of DC volts equal to the RMS of the generated AC sinewave then I'm fine.

Each DC voltage corresponds to a state. I just need to have some AC voltage change when the DC voltage changes. I can put that AC signal into one input of the meter do that I can remotely read it.

Cumbersome yes but I didn't image that a inverter circuit would be that complicated.

Any ideas on a simple DC to AC circuit or something that I can by from an IC manufacturer.

#### tom66

Joined May 9, 2009
2,595
I've seen quite a few digital multimeters which have RS232 ports; would that not work for you? These meters spit out the reading on the display over the connection (usually in plain ASCII) and can do basic datalogging.

#### SgtWookie

Joined Jul 17, 2007
22,220
Here's a fairly simple way to get a pair of opposing 60Hz square waves from a couple of CMOS 555 timers. See the attached. I didn't show the mandatory 0.1uF and 1uF caps across the supply and ground power pins.

#### Attachments

• 119.1 KB Views: 123

#### steveb

Joined Jul 3, 2008
2,436
You could also use any of the available small microprocessors with onboard A/D converters. These will allow you to measure the voltage and then spit out an AC waveform. You can make a nearly perfect wave by using a D/A converter, or use PWM through a low pass filter.

But, I have to say that making an inverter to measure DC voltage is a kludge, particularly if your meter range requires a larger voltage in the range of AC power line levels.

#### Management

Joined Sep 18, 2007
306
Here's a fairly simple way to get a pair of opposing 60Hz square waves from a couple of CMOS 555 timers. See the attached. I didn't show the mandatory 0.1uF and 1uF caps across the supply and ground power pins.
How would I get 60Hz sinewaves from this? Would it just be as simple as some filtering?

#### colinb

Joined Jun 15, 2011
351
How would I get 60Hz sinewaves from this? Would it just be as simple as some filtering?
While waiting for someone to give a better answer, I will say that at least you can't get a sine wave from a square wave using an RC filter.

See Passive Low Pass Filter page for and example of a basic RC low pass filter and look at the end of the page to see what a low pass RC filter does to a square wave.

I did a SPICE simulation just to prove it myself:

Rich (BB code):
ngspice
source rc.net
tran 0.01ms 2ms
plot n0 n1

As an alternative to trying to filter the 555-generated square wave, you might consider another more direct method of creating your sine wave. One easy way to create a high-accuracy sine wave is to use direct digital synthesis (DDS) to generate a sine wave from a low-cost microcontroller and a DAC. I have implemented this and it only takes a few dozen lines of C code to do it.

#### SgtWookie

Joined Jul 17, 2007
22,220
How would I get 60Hz sinewaves from this? Would it just be as simple as some filtering?

To tell you the truth I don't need the RMS voltage of the generated AC signal to equal the DC voltage I want to measure. As long as I know what x amount of DC volts equal to the RMS of the generated AC sinewave then I'm fine.
Did you try building the circuit I posted in reply #5, using CMOS 555 timers? (using CMOS 555 ICs is critical; it WILL NOT work properly with BJT 555 timers.)

Did you try measuring the input DC voltage, and then compare it to the RMS reading of the outputs, measuring A referenced to B?

Did you notice that if the input was 6VDC, that the output measured 6VAC RMS? And the same with the other DC input voltages? (what voltage you supply the circuit using DC, you should read out in VAC RMS.)

If you have done all of that, then why would you want to go to the trouble of making the square wave outputs into sine waves?

On the other hand, if you're not seeing the same VAC RMS out that you're putting in with DC, you should say something.

[eta]
Converting the square wave output to sine waves would require changes that would take it from being a cheap and simple circuit that does just what you need, to a much more complex circuit that would require a separate power supply.

Last edited:

#### #12

Joined Nov 30, 2010
18,216
I'm with wookie. No need to make sine waves. Even if the accuracy sucks horribly, it will suck proportionately, and that is all you really need.

#### SgtWookie

Joined Jul 17, 2007
22,220
I'm with wookie. No need to make sine waves. Even if the accuracy sucks horribly, it will suck proportionately, and that is all you really need.
Actually, unless our OP's meter is loading the output down, I'd expect the VAC RMS to be quite reasonably close to the VDC input.

It isn't a lab instrument, but it should be considerably better than "ballpark".

#### colinb

Joined Jun 15, 2011
351
I guess the assumption is that the meter is measuring the average peak-to-peak voltage and assuming this is the ac rms voltage (valid for true sine waves)?

#### #12

Joined Nov 30, 2010
18,216
Be careful using, "average" and, "peak to peak" in the same sentence. They have different meanings. It can be confusing.

Only a True RMS meter knows the whole truth. All other meters measure the peak voltage, assume it's a sine wave, run it through a voltage divider, and display the results.

#### SgtWookie

Joined Jul 17, 2007
22,220
If his meter is true RMS, it should read correctly.

If it's not a true RMS meter, he'll have to multiply the reading by 1.414 to get close to accurate.

#### colinb

Joined Jun 15, 2011
351
Be careful using, "average" and, "peak to peak" in the same sentence. They have different meanings. It can be confusing. Only a True RMS meter knows the whole truth. All other meters measure the peak voltage, assume it's a sine wave, run it through a voltage divider, and display the results.
Sorry, my use of peak-to-peak was improper--neurons got crossed.

Do they really try to detect and measure the peak, or do the non-true-rms meters just integrate the voltage and take the average? One could test this by feeding a 5 V square wave into the ac meter. If it reads 5 V, then it is doing averaging. If it reads 3.6 V, then it is taking the peak and compensating to estimate rms.

On second thought, if the meter were taking the mean voltage through integration, it would then be a true-rms meter, would it not? So cheap meters must be using peak voltage and scaling it.

#### #12

Joined Nov 30, 2010
18,216