1
\$\begingroup\$

I am looking at the MC1496 IC datasheet provided by ON here.

Figures 1 and 3 show that you can make an AM or DSB-SC signal using this circuit. However, the two test circuits listed on the datasheet for the two are the same circuit:

Typical AM vs Typical DSB SC

My question is, how can I implement one versus the other? I have followed this circuit schematic, intended to create a DSC-SC circuit:

The Circuit I Actually built.

When using the message signal as 8mVpp at 200 Hz, and the carrier as 2Vpp at 20kHz, the resulting output is an AM signal instead of a DSB-SC signal.

Actual Output

Does anyone have any suggestions as to why this might be happening, and if so, have any suggestions for implementing a DSB-SC circuit instead of an AM circuit using this chip?

\$\endgroup\$
11
  • \$\begingroup\$ Look, if you add a DC signal to your "message" before multiplication, then you have AM. Otherwise it is DSB. I guess I'm not sure about the problem, just yet. I'll have to read better. \$\endgroup\$
    – jonk
    Commented Jun 25, 2020 at 4:56
  • 1
    \$\begingroup\$ Is your "null adjustment" actually nulled, even down at the millivolt range? (Try 0mV message modulation. The carrier ideally should be zero output.) Hmm, maybe the chip doesn't entirely suppress carrier, and the bit of leakage is being AM modulated. In that case you need to feed it FAR higher "message" signal. Perhaps 200mV not 8mV. Also, the diff betw AM modulation and DSB is about whether the Null setting (during zero message signal) is set for zero carrier out, or set for 50% carrier out. \$\endgroup\$
    – wbeaty
    Commented Jun 25, 2020 at 5:24
  • \$\begingroup\$ Look for places where 4mV of DC offset may have crept in. \$\endgroup\$
    – user16324
    Commented Jun 25, 2020 at 9:21
  • \$\begingroup\$ @wbeaty Thank you for your clarifications on the differences between AM and DSB-SC when using this chip. Last night, after I posted this message, I noticed that I did not have the null adjustment completely set to zero, but after manipulating it (with a 10K potentiometer) down to the smallest amplitude possible, this issue was still occurring. When I'm able, I'll test out your suggestion of using a much larger message signal, and see if anything changes. \$\endgroup\$
    – kbyrum
    Commented Jun 25, 2020 at 12:15
  • 1
    \$\begingroup\$ Your circuit shows message input with DC-coupling. Any DC-offset from your message-signal voltage source will disrupt R2's balance setting. For DSB-SC operation, you drive 20 kHz into carrier port, and adjust balance to minimize carrier amplitude at the output (with zero message signal going in). That minimum null-point may be hard to achieve...as with most "nulls", it is very sharp. \$\endgroup\$
    – glen_geek
    Commented Jun 25, 2020 at 13:06

3 Answers 3

1
\$\begingroup\$

To check:

  • be sure you DC_block the modulating input

  • alter the gain, by using 100 ohms between pins 2 and pin 3

  • if not over a Ground plane with good VDD bypassing, then phase_shifts vi the power rail might upset the internal charge cancellations that produce the carrier suppression

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

The two test circuits are identical because they differ in the position of the "carrier null" potentiometer.

Paraphrasing from comments: the message input is also DC coupled. Any input DC offset is added to R2's balance setting. For DSB-SC operation, you minimize message DC bias (whether by AC coupling or otherwise), and adjust R2 to minimize carrier amplitude at the output when message is zero. For AM, simply add an offset -- whether from the input or via R2.


I have a function generator with exactly this circuit (well, I haven't checked, but functionally at least), which has a "carrier null" control on the front panel, for its AM mode. It does have a "suppressed carrier" button, which basically shifts the range of the null control; the ranges overlap I think, so it's not a particularly useful button, but it does allow the range to be narrower (easier to set).

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

OP has correctly raised red flags upon seeing LABVIEW display showing a 100%-modulated 20kHz carrier: an ideal AM output. The question posed:

Does anyone have any suggestions as to why this might be happening, and if so, have any suggestions for implementing a DSB-SC circuit instead of an AM circuit using this chip?

Had the carrier null adjustment been done, what should have appeared is a DSB-SC (suppressed carrier) waveform. A DC offset is the cause, but its source is not clear.

  • 200 Hz signal generator (8mV amplitude) may have added a DC offset - it is DC-coupled
  • Carrier null potentiometer may require careful adjustment

It is odd that this setup creates such an ideal AM waveform with 100% modulation index. I can only suggest that 8mV amplitude has been chosen to perfectly match an 8mV DC offset, resulting in the AM display rather than the expected DSB-SC display. A higher amplitude should overmodulate...moving toward double sideband, but not yet with a fully-suppressed carrier.


One might AC-couple the 200 Hz modulating signal in, so that any DC offset from this signal source is blocked. With no 200 Hz signal applied, the carrier null pot is carefully adjusted while monitoring the modulator's output - adjust for minimum carrier. When 200 Hz modulating signal is once again applied to the AC-coupled input, you should see DSB-SC output, not AM.
With gain resistor R10 set to 100 ohms (rather than 1k in data sheet), modulation input has 10X more gain, making carrier offset null adjustment more finicky.

If AM is wanted instead, the carrier null pot is adjusted to one side (either side is fine). OP's R1 (10k) and/or R2 (10k) may have to be reduced in value to get enough offset range for AM.

\$\endgroup\$

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.