Adding a signal meter to a broadcast receiver

Status
Not open for further replies.

dr pepper

Well-Known Member
Most Helpful Member
I have a broadcast am/fm/lw receiver I use for 162kc time code reception, they used to broadcast audio but now its just a psk modulated carrier so its hard to tune with a cheap radio.
I found the agc filter cap for the am side, it has 1v with no signal and 0.8v with a moderate strength station, as expected the higher signal the lower the voltage and therefore gain.
I was wondering how to roughly calibrate either a meter movement or a Lm3915 as an S meter, would the max signal be about 0.6v?, as f ar as I can tell the agc bus goes to the mixer/osc and the first of 2 If stages.
this could be done with an expensive yaesu however this is what I have.
 
I was wondering how to roughly calibrate either a meter movement or a Lm3915 as an S meter
I would be inclined to use a moving coil meter which has infinite resolution, rather than an LM3915 which has just 10 discrete level indications. Fine tuning using such a meter is quite frustrating. (Been there, done that, and would prefer not to do it again.)

In terms of "calibration", unless you have a signal generator with an accurate output attenuator, probably the best that you can do is to find the strongest signal in the frequency range which interests you and make the meter move to about 95% of full scale when you receive this signal. And of course set the meter to read 0% when there is no signal.

I found the agc filter cap for the am side, it has 1v with no signal and 0.8v with a moderate strength station, as expected the higher signal the lower the voltage and therefore gain.
Using a couple of op-amps, you can pick off the AGC voltage and make some offset and scaling so that the meter reads from zero to 95% or whatever you choose.

Here is something which I made a few months ago for an Elecraft K2 transceiver, which has the dreaded 10 LED "blobs" for an S Meter.
The K2 has an AGC voltage which increases with increasing signal strength.
This circuit of mine has two sensitivity ranges, because a wanted a position where I could have a meter sensitivity of just 2 or 3dB for zero to fullscale.
In the low sensitivity position it has the normal sensitivity of 0 = no signal, to full scale = 2 or 3mV at the antenna (ie very strong signal).

JimB

 
I've lashed up something similar, an lm358 but I connected the agc bus to the inverting i/p, biased the non inverting input to 1v the zero signal voltage of the agc bus, it works.
But I found the op amp wont go quite to ground, so I'm going to make the thing a summing amp and use the other i/p to offset the agc voltage so its starts at the mid rail voltage, so the meter see's 0v with no sig.
Point taken about the analogue meter, I think I have something knocking around, it would have to be an edgewise to get it in. I have a sony comms rx more of a fancy radio, it has 6 leds that are on or off, a lot of japanese or chinese stuff work like this, at least the Lm3915 cross fades the leds.
I was thinking for calibration to connect my uncalibrated sig gen to my 'posh' comms receiver adjust for 0db at the freq of interest, then connect the cheapie rx'er and adjust the pot for 0db, might have to do that for no sig and 0db to make sure the zero and span are fairly close.
 
The AGC volts on the K2 run from 3.3v with no signal to 4.45v with an input of -50dBm (about 1mV), so there was no problems with rail to rail capabilities of the inputs of the op-amp.

But for the output, I used the second half of the dual op-amp to provide an offset of about 1v for the -ve side of the moving coil meter, so that the op-amp output is well within the available range.

JimB
 
Makes sense.
Theres a possibility I'll be connecting this to a uP so I was trying to get the meter to be ground referenced, I spose I could use a different ground ref for the atod.
After dark other stations come wandering in on 162 kc, my comms rx and kiwi online sdrs's in the area show theres no audio broadcast near, the closest being 153kc, either I'm getting some weird intermod or the cheap rx is on the wide side, I'm going to try realigning the set, and maybe improve its selectivity.
Failing this I'll add a bfo 2 or 3 kc of 162 & redesign the detector circuit to work at audio freqs, with a comms rx I get a good signal on cw and you can easily hear the psk, should be able to inject the bfo with a bit of wire near the ferrite rod.
 
Last edited:
you could (if you have a stable VFO oscillator) do a direct conversion receiver, which CAN be as simple as 4 diodes, 2 RF transformers and a potentiometer (which is a balanced modulator/demodulator)(and maybe a tuned RF amp for better reception). you would then tune the VFO to the frequency you want to receive, and take the output from the second transformer, or, even simpler, is the balanced detector in this receiver that uses no rf transformers: http://www.ke3ij.com/DC-80.htm all you have to do is modify the input filter and the oscillator for 162khz. gee, i'll have to try that... i can use my HackRF as the local oscillator...

i would use germanium diodes in place of the silicon ones... actually i have an RF generator sitting around gathering dust, along with a frequency counter... i might try this without the HackRF...

looks as if it can be built in about an hour or less
 
Last edited:
Interesting design with diode switches, reminds me of an iq receiver.
I'll bear that in mind if this old rx wont play.
I have some very nice salvaged geranium diodes, to18's I think, like bc108's.
That said running 164kc from my sig gen along a wire next to the ferrite produces a loud 2kc tone, with very little audible interefence, Just experimenting adding a 2kc wide ceramic 455kc filter after the second If, you can tell the audio bandwidth is narrower and a lot of the mush goes away, distorts like mad though on loud stations, I have too much gain in the passband.
Decisions, carry on with this old pile or build new, I'll wait to see how you get on Jed however although this circuit is more complex its 'interesting' an xtal Lo and an If bfo would make a respectable rx, that tracking filter wont be an issue for just one station freq.
https://sites.google.com/site/radiolabrio/short-wave-ne602
 
Last edited:
Found this, a very nifty circuit, highly selectable and not too complex, might combine the above's cascode with something like this, obviously jiggle the freq's around.
**broken link removed**
Edit: maybe not change the freq of the filter, the impedance of anything lower than 8mc crystal wise goes right up, so I'll have to shift up to 8mc.
 
Last edited:
i've owned radios where i used an external 455khz oscillator as a BFO, and in the early attempts, it was difficult to maintain the right balance between incoming signal and the BFO, so what i did later was use a potentiometer for the BFO output level, which helped. then i got a Knight R-195 that had the BFO and a product detector, and that worked much better. now that i have real ham gear, and SDRs, i haven't had to mess with such kluges. but i might build that direct conversion circuit to see how it performs. i always had mixed results with the diode ring demodulator, i think from not having quite enough signal out from the generator to forward bias the diodes.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…