1
\$\begingroup\$

I would like to measure the efficiency of an HF (3-30 MHz) class C power amplifier. In order to do this, I have to accurately measure AC+DC RMS current in the order of 1 A or so. Most instruments are not designed to measure AC current above ~100 kHz. One option I'm aware of is to connect the power supply through a 1 Ohm resistor and measure Vrms on this resistor with an oscilloscope and a differential probe. The probe is needed, because otherwise the PSU will be shorted to ground through the oscilloscope. Unfortunately, the differential probe I have is a high voltage and noisy one. It shows ~200 mVrms even if there is no current.

How would you solve a task like this? Should I look for a low voltage differential probe, a special current probe, or some other tool? Is it difficult to home-build a suitable probe? Or maybe there is a completely different way to solve the task?

\$\endgroup\$
1
  • \$\begingroup\$ Where would the current flow if you're not measuring it? Is it really current you're interested in, or is that just a "proxy" for something else? \$\endgroup\$ Aug 16 at 13:22
3
\$\begingroup\$

How would you solve a task like this?

It's the average current into the amplifier's power supply connection multiplied by the DC voltage at that pin that determines the true power consumed by the amplifier when driving an AC signal to its load.

In order to do this, I have to accurately measure AC+DC RMS current in the order of 1 A or so.

If you are measuring the current into the amplifier's power connection then you are only interested in the DC (average) current (if calculating power into the circuit as supplied from the DC supply voltage).

Reason: DC volts x AC amps has an average level that is zero and, it is average power you are interested in. DC volts x DC (average) amps does not have an average level of zero i.e. it represents the true power taken by the device.

\$\endgroup\$
1
\$\begingroup\$

Measuring DC power is relatively easy with a current shunt and LPF or DSO measuring Avg. or RMS if there is ripple.

Measuring differential AC power to the load does required balanced probes with very low inductance ground and probe connections < 1= cm with test pins and calibrated probes on one signal to get a flat line in A-B mode removing tip and ground clip and making very short connections to sig&gnd. You may even have to twist the probe cables of equal length to the DSO. If the probes are still inadequate, use AC coupling caps to two 50 ohm cables terminated with some low R that draws some or all of the load.

If you can do this your noise levels should drop at least an order of magnitude and if balanced well, a CMRR of >> 40 dB.

Then you can compute efficiency.

\$\endgroup\$
2
  • \$\begingroup\$ Measuring the RMS when ripple is present is wrong. For a stable DC supply, you measure average current. \$\endgroup\$
    – Andy aka
    Aug 17 at 8:08
  • \$\begingroup\$ @Andyaka right but that’s the easy part when SNR is poor. \$\endgroup\$ Aug 17 at 13:10
0
\$\begingroup\$

So I found an answer in the recent video on FesZ Electronics YouTube channel. As it turns out, you can make a decent AC current probe simply by winding several turns on a suitable ferrite core and soldering the ends of the windings to the SMA/BNC connector.

Personally, I used 10 turns of 0.9 mm solid copper wire on Amidon FT50-43 ferrite core. When loaded to 50 Ohms, this works as a ~4.75 mV/mA current probe with a very flat frequency response in the 100 kHz - 150 Mhz range. To improve the sensitivity an LNA can be used, like those that are cheaply available on eBay.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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