Downconverter Mixer Design

[Chunx]

Hey guys, Im trying to use a 100kHz downconverter for an AM demodulator. Ive got a circuit diagram for a 900MHz mixer, but Im using a 100kHz signal (for both the RF and LO inputs) and I'm not sure how to re-calculate the capacitor and inductor values to accommodate a 100kHz single ended signal, so any help would be greatly appreciated. If there are any better ways to do this please let me know!

 

[Fish4Fun]

Chunx,

Welcome to the forum! RF is REALLY not in my wheelhouse, I am sure others will be able to help, but "off-the-cuff", I thought AM stood for "Amplitude Modulation" while FM stood for "Frequency Modulation".....I have never done more than take a precursory look @ RF in general, and the things I DONT KNOW would fill volumes, but it seems to me you might have trouble using an FM demodulator to Demodulate an AM signal...I don't think it is going to be a simple matter of recalculating the passive filter values....but then again maybe I am way off-base.

Good Luck!

Fish

 

[davenn]

Hi Chunx

welcome to EP

you cant mix 100kHz with 100kHz and end up with something useable as the result out of the mixer will be 0 (zero)

what are you actually trying to achieve ?

Dave

 

[Laplace]

... a 100kHz downconverter for an AM demodulator ... using a 100kHz signal (for both the RF and LO inputs)

With that description one can only be designing a coherent demodulator. But in order to avoid the unusable output alluded to by davenn, synchronous AM demodulation requires the LO to be an exact duplicate of the carrier in both frequency and phase, which can at times be difficult to achieve when there is no carrier present such as in single sideband (SSB). So what are you trying to achieve? Nevertheless, impedance matching at 100KHz is not quite so stringent as at 900MHz. You could have selected a lower-frequency example from the datasheet.

 

[Chunx]

Chunx,

Welcome to the forum! RF is REALLY not in my wheelhouse, I am sure others will be able to help, but "off-the-cuff", I thought AM stood for "Amplitude Modulation" while FM stood for "Frequency Modulation".....I have never done more than take a precursory look @ RF in general, and the things I DONT KNOW would fill volumes, but it seems to me you might have trouble using an FM demodulator to Demodulate an AM signal...I don't think it is going to be a simple matter of recalculating the passive filter values....but then again maybe I am way off-base.

Good Luck!

Fish

Yeah, it looks more an FM demodulator to me! The idea is to mix a 100khz message with the original 100khz carrier (the actual message envelope is about 1hz). I would have thought to use an envelope detector, but Im running of a paper which says use a mixer so the phases will need to be perfectly aligned at 180... I think!

 

[Chunx]

Hi Chunx

welcome to EP

you cant mix 100kHz with 100kHz and end up with something useable as the result out of the mixer will be 0 (zero)

what are you actually trying to achieve ?

Dave

The RF has a 1Hz message envelope, thats what Im trying to extract

 

[Chunx]

With that description one can only be designing a coherent demodulator. But in order to avoid the unusable output alluded to by davenn, synchronous AM demodulation requires the LO to be an exact duplicate of the carrier in both frequency and phase, which can at times be difficult to achieve when there is no carrier present such as in single sideband (SSB). So what are you trying to achieve? Nevertheless, impedance matching at 100KHz is not quite so stringent as at 900MHz. You could have selected a lower-frequency example from the datasheet.

View attachment 15210

Thanks heaps! Im mixing the 100khz signal with a 1Hz message envelope with the original 100k carrier. I dont think Im going to worry about impedace matching anymore... should still work I hope. Im probably grounding both inverting inputs. Both signals are coming straight from Instrumentation amp

 

[Chunx]

I should clarify, the signals Im measuring arent actually RF. Im building an ecg by running a small AC current through the wrist and measuring the change in impedance. The signal that comes out is an AM modulated signal which Im mixing with the original carrier

 

[Laplace]

The first thing I would do is an experiment to measure the carrier phase delay through wrist conduction. If the phase delay is significant enough, then the carrier oscillator signal may need to be phase shifted a similar amount before going into the mixer. If there is a significant variablility in phase delay from one wrist to another, then it becomes an interesting problem.

 

[Chunx]

I'm going to test the circuit as soon as the parts come in! If the signals are significantly out of phase, is there a simple way to shift the phase linearly?