0
\$\begingroup\$

Disclaimer: I'm computer scientist

I have a sinus wave generator (handyscope H3-5) from TiePie but the output power is insufficient for my application. The specification is here. I need at least 60W. I would like to know if it is OK to connect the generator to the input line of an audio amplifier. I need only a signal below 20kHz.

The amplifier is an SMSL SA 98E. I only have a french data sheet. The data sheet states that the input line is 400mV-2V.

I'm not sure what it means.

Also I know that the generator is 50Ohms. Do I have to care about that ?

I plan to buy an RCA<->BNC cable. These are sold for video connections. I think they are 70Ohms. Would it be OK ?

\$\endgroup\$
14
  • \$\begingroup\$ Disclaimer: I'm computer scientist All creatures are welcome here. \$\endgroup\$
    – Marcus Müller
    Commented Jan 6, 2021 at 14:57
  • \$\begingroup\$ Do I have to care about that ? No, you can simply ignore the 50 ohms and 70 Ohms as those numbers will only be relevant when you use frequencies that are much higher than 20 kHz. A BNC to RCA cable will very likely just work. Do realize that the amplifier is a class D amplifier so it has a switching output stage. For using it with a loudspeaker or a resistor, that does not matter. If you use it to drive something else there might be issues. \$\endgroup\$
    – Bimpelrekkie
    Commented Jan 6, 2021 at 15:00
  • \$\begingroup\$ @Bimpelrekkie I didn't know that. I want to connect a 4Ohms coil to it. Would that be a problem ? \$\endgroup\$
    – chmike
    Commented Jan 6, 2021 at 15:01
  • \$\begingroup\$ @chmike that sounds like essentially a big mass, that will "filter out" all the higher-frequency components that might arise from the class D amplification. What is that coil used for? \$\endgroup\$
    – Marcus Müller
    Commented Jan 6, 2021 at 15:04
  • \$\begingroup\$ @MarcusMüller the coil is use to generate a varying magnetic field to interact with a liquid substance in its core. I can't reveal more of the application. \$\endgroup\$
    – chmike
    Commented Jan 6, 2021 at 15:07

4 Answers 4

Reset to default
3
\$\begingroup\$

The data sheet states that the input line is 400mV-2V.

That is probably OK; I'd assume you can easily set your output to that amplitude range.

Also I know that the generator is 50Ohms. Do I have to care about that ?

Not in this case: the input impedance of an audio amplifier will be very high, so that the source doesn't have to drive much load (i.e. be able to supply much curent).

And, at 20 kHz you don't have to care about wave impedances etc: your cable is much, much smaller than the wavelength of a 20 kHz electromagnetic wave.

So, this will work.

Note that your generator is absolutely overkill.

A sound card would do just as well – these are audio frequencies, after all!

Also, everything but the worst sound cards have astonishingly high fidelity, so that's often an attractive choice. You'll want to be a little careful when using a sound card that it has (or you add) good anti-imaging filtering (i.e. to suppress harmonics that arise from the discrete-time sampling).

\$\endgroup\$
9
  • \$\begingroup\$ Thank you very much. I already have the generator and I would only have to plug it. With a sound card, I would need a software that I currently don't have to generate the sinus. But I plan to use a sound card so that I can change frequency by software at will. \$\endgroup\$
    – chmike
    Commented Jan 6, 2021 at 15:05
  • 1
    \$\begingroup\$ You can generate any waveforms you want with any programming language and then play them back. You can also generate or open waveforms in Audacity and play them with it. Many mobile phones have app stores that include signal generators too. \$\endgroup\$
    – Justme
    Commented Jan 6, 2021 at 15:09
  • \$\begingroup\$ @chmike: Pure Data is just about the easiest software to use for generating signals through the sound card. It is free (in all senses of the word.) It is also available for multiple operating systems (Windows, Linux, Apple.) \$\endgroup\$
    – JRE
    Commented Jan 6, 2021 at 15:15
  • \$\begingroup\$ @JRE I DIDN'T EVEN KNOW THAT! And I should know things like that. \$\endgroup\$
    – Marcus Müller
    Commented Jan 6, 2021 at 15:16
  • 1
    \$\begingroup\$ Pure Data is a few years older than GNU Radio, which I also use for audio signal generation. Pure data is easier to use to generate signals, GNU Radio is easier to use if you need to generate plots. \$\endgroup\$
    – JRE
    Commented Jan 6, 2021 at 15:19
2
\$\begingroup\$

OK to connect the generator to the input line of an audio amplifier?

Yes, this is done all the time.

The cable impedance is not important at audio frequencies.

Set your function generator output level to about 0.5 V p-p.

What are you connecting to the output of the amplifier? Here is where you need to be more concerned. Audio amplifiers are not tolerant to low impedance, 4-8 ohms is usually the minimum, look at the specs.

If you are driving speakers, beware. A speaker may not tolerate a lot of energy at one frequency.

\$\endgroup\$
2
\$\begingroup\$

Yes you will be fine - but beware that very LOW frequencies (less than 20Hz or so) may cause problems. Audio amps (good ones) block DC and frequencies below about 10-20Hz, for good reason : that stuff can kill speakers at high levels.

What are you actually driving? what's the load, and what's the application?

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Thank you for answering. The load is a 4 Ohms coil. It is for an experiment. I put a liquid in the core of the coil. I can't reveal more \$\endgroup\$
    – chmike
    Commented Jan 6, 2021 at 15:26
2
\$\begingroup\$

All the other answers have adequately covered your question. I'm going to add a reminder that audio amplifiers are not precision equipment.

Your amplifier will be able to faithfully reproduce the signal you give it, as long as you keep your signal within the input limits of the amplifier and as long as you don't "crank up the volume" to the point that the output clips (distorts.)

If the output level is in any way critical to your experiment, then you will need to measure the output level. Your output level is related to the input level by the amplification (gain) of your amplifier, and by the setting of the volume control knob.

The volume control isn't calibrated in any way - you can't rely on it for absolutely reproducible results.

Since you don't (can't) say what the experiment is about, no one can even begin to guess how critical the output power is.

If it's critical, measure it.

As a value for repeating your own steps, the output voltage (peak to peak or RMS) would be a good starting point. You'll need either an oscilloscope or an AC voltmeter to measure it. If you use an AC voltmeter, you'll need one that covers the entire audio range. Common hardware store voltmeters cover just the typical household AC line frequencies (50 or 60 Hz.) AC voltmeters that cover up to 20kHz can be expensive.

For whatever paper you write about your results, you may need to include to actual power into the coil, and maybe the waveform as actually fed into the coil.

\$\endgroup\$
2
  • \$\begingroup\$ Thank you for the advise. I’ll then use an oscilloscope to measure the voltage. How do I determine the power from it ? Am I required to measure the current ? Can’t it be deduced from V and the impedance of the coil ? \$\endgroup\$
    – chmike
    Commented Jan 6, 2021 at 16:03
  • \$\begingroup\$ You can calculate power from impedance and voltage - if you know the impedance accurately. Impedance is often not as simple as you might think - a coil can have large variations just over the audio range. There are AC power meters out there that cover the audio range. Expect them to be expensive. You might be able to rent them. \$\endgroup\$
    – JRE
    Commented Jan 6, 2021 at 16:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.