SSB Carrier Supression

[xanadunow]

Hi lefty, a closer look at the licence you hold in your hand dos not prohibit us from using A3 - it is on the list.

Taking it many years back I do remember well ham community broadcasts on 80m done on AM and I would not like them in any other way even now. But I do make distinction between the broadcast and the communication in ham radio.

Having a chat on AM between friends is a legal and acceptable, especially if we turn off our QRO stages remembering that we are not the only ones on the band and day for me on 80m is a night for someone else and possible DX conditions.

But when chasing DX or trying to win in a contest you will, and I will too, resolve to the most efficient manner to achieve that goal and AM just won't cut it..

 

[xanadunow]

Yes we were, but now we are not. And, there is nothing to it. I do love the high quality of the sound in my lounge as much as I love the high efficiency of my ham gear in my shack.

 

[Mikebits]

It would seem to me that L1 and C15 would shunt out your xtal by presenting a very low impedance. What value inductance have you figured for L1? I am guessing your oscillator is just using L2 and the Cap, but heck if you say it works who am I to argue?

 

[Space Varmint]

Ability to hear any range of (audio) frequencies between 20Hz and 20kHz is very much an individual matter of a human and is rarely attributed to a particular "hobby" as implied in the message by uncle above. If anything - the age, the listening habits and perhaps, the occupation (namely the protective gear for the ears) - are the major factors in the deterioration of frequency boundries we can hear as individuals during the span of our life.
But it is well outside this particular thread, we were talking about SSB here.

Hey there xanadunow Welcome aboard brother! Don't worry about audioguru. I've been knowing him a couple/few years through this and another website. I like the crotchety old man anyway...lol. Hey, he is definitely king pin in audio, but the poor guy is lacking in rf so he pokes fun at me........fill in the blanks.

Anyway, in this thread I'm discussing an SSB transmitter. I was getting ready to put in an MC1496 chip, but think I will back down. Most all of the spec sheets show a split power source. Would like better documentation on a single power supply before I would attempt it.

Anyway, it is possible to produce SSB with just a very sharp filter. So what I think I will do is go back to the diodes. Maybe these germanium diodes I just picked up will work better.

 

[Space Varmint]

It would seem to me that L1 and C15 would shunt out your xtal by presenting a very low impedance. What value inductance have you figured for L1? I am guessing your oscillator is just using L2 and the Cap, but heck if you say it works who am I to argue?

Well, to be honest with you Mike, I never had a problem warping the frequency up. Wish that was the direction I needed to go. Would have saved me allot of time. ANyway, you may be onto something, because like I said, if the caps are set correctly, then I cannot reach the crystal's center frequency. If I touch the shaft on the cap the frequency will jump around unlike if I'm on the high side of the crystal. The thing is, is it passes all the tests for what is needed. It does hold the frequency and will fire up on the same frequency it was set to, and I took a space heater and put it right over the circuit for a couple of minutes and it held. I even ran it over night and when I woke up it was right on frequency. It is also jitter free and has a very nice output signal. I tend to think by the way it is behaving, that what is going on is, it is set up to remain a set amount of kilohertz from the crystal's frequency. This is really the hardest thing to do because the desired offset is only 1.5KHz.

I would say, it is safe to move on.

 

[JimB]

Pondering about the power consumption was the leading factor to discovery that in the AM modulation - 50% of the transmitter's power is consumed by the "carrier" frequency and 25% for the "side-band" each. There are two of them, the upper (USB) and the lower (LSB).

It was also discovered, that for legible reception we do not need a carrier (50% of power saving) and out of the two - only one side band is required (another 25% of power saving). It meant, that the whole whooping 100% of transmitter's power can be dedicated to transmit the only thing required for a legible reception of the voice - a ONE Singe Side Band (hence the SSB). USB or LSB - to be more exact.

Not quite right, you are actually understating the advantage of SSB over AM.

In a 100% modulated AM signal, the TOTAL power in the sidebands is half the power of the carrier.
So the power in one sideband is one sixth of the total (carrier + two sidebands).

JimB

 

[flat5]

"I was getting ready to put in an MC1496 chip, but think I will back down. Most all of the spec sheets show a split power source."

I used it with a single 12v supply.
I got the circuit from either an ARRL Handbook or an issue of Ham Radio of the 1970s. Just maybe the 60s.

 

[Nigel Goodwin]

"I was getting ready to put in an MC1496 chip, but think I will back down. Most all of the spec sheets show a split power source."

I used it with a single 12v supply.
I got the circuit from either an ARRL Handbook or an issue of Ham Radio of the 1970s. Just maybe the 60s.

Likewise, I used an MC1496 on a single 12V supply, many years ago when I built an SSB receiver (never got round to making it into a transceiver, as I never bothered taking my morse test).

 

[xanadunow]

[quoting JimB]:
"So the power in one sideband is one sixth of the total (carrier + two sidebands)"

Maybe so.. it is not what have been tought and belived all these years.. 100% divided into quarters means that every quarter takes the 25%.. You have committed 50% to the side carriers, meaning 25% each.. this makes sense and this is what I was stating..

But suddenly, you say - the advantage is greater than that. The more, the better - but bring some evidence to explain 1/6th of the power against the 1/4 as I have stated..

What we have now is 50% of the power in the carrier, and 1/6+/1/6 i the two sidebands.. that mkes together 3/6 + 1/6 + 1/6 = 5/6 of the total power.. Where is the other 1/6?

If I am to learn, now is a very good time. I am very open to suggestions.

Whilist it is understandable by me; the 50% of the carrier can not be fully supressed (and it works against my statement), even on the assumption that the suppression is ideal (i.e. 50%), the ramaining 50% must be devided amongst what remains (the two carriers) equally and it means 25% each..

It is the output power, not the one measured at the power meter registered on the household electricity meter provided by the electricity company.

I am so much puzzled by the 1/6th statement.. that I ask for an explanation. If it is the truth, then I must have missed something along the lines and my certificate may need to be revoked. Mayby not, after all, I am the CW man

I am eager to learn.. please follow up with more information Jim..

Regards,
xanadunow

 

[audioguru]

The datasheet for the MC1496 shows most of the circuits with a dual polarity supply because both inputs are biased at 0V. The Balanced Modulator circuit shows a single supply and resistors biasing both inputs positive a few volts.

 

[JimB]

I am eager to learn.. please follow up with more information Jim..
xanadunow

OK Lets try a quick and dirty explanation first.
To 100% modulate a 100watt carrier, you need 50watts of audio, (assuming high level plate and screen modulation for us old thermionic types).
So we have a total of 150 watts, 100w carrier and 50w in the sidebands. The sidebands are identical so there is 25w in each sideband.

25/150 = 1/6

A bit more sophisticated explanation.
Looking with an oscilloscope, an un-modulated carrier has a peak voltage V.
When modulated to 100%, the maximum peak voltage is 2V and the minimum peak voltage is 0.
The peak voltage varies from 0 to 2V at the modulation rate.
This variation is due to the voltages of the two sidebands adding to and subtracting from the carrier.
The sidebands are identical in amplitude so the voltage of each sideband is V/2.

If the voltage is halved the power must be divided by 4.

So each sideband has 1/4 the power of the carrier.
If the carrier power is 100W the power in each sideband is 25W, to tal power =100 + 25 + 25 = 150W

25/150 = 1/6

OK XanaDoesKnowNow ?


JimB

 

[xanadunow]

I am very happy now JimB. You have added the audio power to the RF power. It is not such a bad idea, but it is not the way I was tought. In the overall balance of power it should be taken into account indeed. Thank you. What was I thinking of Just as well I did not have my 500W toaster going at the same time; it would complicate the matter even further I suppose

xanadunow

 

[Leftyretro]

OK Lets try a quick and dirty explanation first.
To 100% modulate a 100watt carrier, you need 50watts of audio, (assuming high level plate and screen modulation for us old thermionic types).
So we have a total of 150 watts, 100w carrier and 50w in the sidebands. The sidebands are identical so there is 25w in each sideband.

25/150 = 1/6

A bit more sophisticated explanation.
Looking with an oscilloscope, an un-modulated carrier has a peak voltage V.
When modulated to 100%, the maximum peak voltage is 2V and the minimum peak voltage is 0.
The peak voltage varies from 0 to 2V at the modulation rate.
This variation is due to the voltages of the two sidebands adding to and subtracting from the carrier.
The sidebands are identical in amplitude so the voltage of each sideband is V/2.

If the voltage is halved the power must be divided by 4.

So each sideband has 1/4 the power of the carrier.
If the carrier power is 100W the power in each sideband is 25W, to tal power =100 + 25 + 25 = 150W

25/150 = 1/6

OK XanaDoesKnowNow ?


JimB

I seem to recall in the old days that the simplest expression of the power efficiency of SSB Vs AM was that there was a +9db improvement in the S/N ratio at the receiver's location. I assume if correct, that part of the improvement is based on the decreased bandwidth used in the receiver optimized for SSB.

Not sure if the 9db improvement was correct or not but it sure allowed Collins Co. to sell a lot of new radios to the US military in a very short time frame.

Lefty

 

[Nigel Goodwin]

I am very happy now JimB. You have added the audio power to the RF power. It is not such a bad idea, but it is not the way I was tought.

He's only describing one particular method of producing AM, high level modulation using an audio power amplifier and a transformer.

For the point of this discussion it's confusing things a lot

An alternative method (and the modern radio amateur way) is to do it with a linear amplifier like SSB does, using the same modulator and everything else. You switch the modulator to NOT cancel the carrier, and the filtering NOT to remove the other sideband, and you have a perfectly fine AM modulation that goes through the linear amplifer stages just as SSB does. This is the method which you should compare to SSB, and is what you already understood.

 

[xanadunow]

Thanks Nigel, this is why I did mention the toaster but I guess, I was to subtle. The first ever SSB rig I had could do exactly what you have described but I must lower my head because the broadcast AM is the true standard to compare against and I have actually missed that in otherwise true and down to earth explanation. As JimB would say - "the pseudo AM won't just cut it"

Regards,
xanadunow

 

[Space Varmint]

He's only describing one particular method of producing AM, high level modulation using an audio power amplifier and a transformer.

For the point of this discussion it's confusing things a lot

An alternative method (and the modern radio amateur way) is to do it with a linear amplifier like SSB does, using the same modulator and everything else. You switch the modulator to NOT cancel the carrier, and the filtering NOT to remove the other sideband, and you have a perfectly fine AM modulation that goes through the linear amplifer stages just as SSB does. This is the method which you should compare to SSB, and is what you already understood.

Yeah but your power output will be the same. It may take more power to produce AM high level modulation, but the net result in the final is exactly the same power dissipation.

 

[Space Varmint]

The datasheet for the MC1496 shows most of the circuits with a dual polarity supply because both inputs are biased at 0V. The Balanced Modulator circuit shows a single supply and resistors biasing both inputs positive a few volts.

Well, I just don't understand the chip well enough. I ended up with about 6 data sheets and app notes. What a bunch of _____. The ony thing I need is some suppressed carrier and I want it ballanced. The fact that you don't even need a balanced modulator to produce SSB and can be done with a filter alone, is enough for me to realize what I don't get from the balanced mod. I can make up for in the filter. In fact I think I'm gonna follow up this filter with a partial ladder network. I think that could be a big part of where the residual carrier was coming from.

 

[Leftyretro]

Well, I just don't understand the chip well enough. I ended up with about 6 data sheets and app notes. What a bunch of _____. The ony thing I need is some suppressed carrier and I want it ballanced. The fact that you don't even need a balanced modulator to produce SSB and can be done with a filter alone, is enough for me to realize what I don't get from the balanced mod. I can make up for in the filter. In fact I think I'm gonna follow up this filter with a partial ladder network. I think that could be a big part of where the residual carrier was coming from.

Well the MC1496 is a pretty adaptable device but with that comes some complexity in interfacing with it. As you already stated the data sheet tends to show it's use with a bi-polar power supply and differential input and output, all of which tend to help with the main objective of having the best balance possible. However many adaptations have shown it can be used with single polarity power and single ended signals but it takes more coupling caps and bias arrangements. It's been a long time sense I've looked at circuits using that chip but it can be made to work well, it was a very popular device for that application.

Not sure how well one can generate a SSB signal without a balanced mixer as that puts a lot of the burden on how vertical the IF filter (crystal or mechanical) bandpass shape is. 40+db of carrier suppression was a common goal for many designs. Any significant carrier frequency not removed can result in a annoying whistle at the receiver's location.

Lefty

 

[audioguru]

The balanced modulator cancels the carrier. I don't think a filter will remove enough of the carrier without also removing some of the wanted sideband.
The filter selects the wanted sideband and attenuates the unwanted sideband.

 

[Leftyretro]

The balanced modulator cancels the carrier. I don't think a filter will remove enough of the carrier without also removing some of the wanted sideband.
The filter selects the wanted sideband and attenuates the unwanted sideband.

That was my point, although there are extremly sharp (and expensive)bandpass filters using many stages that can have almost straight "brick wall" passband shape. However I've never seen a SSB modulator/demodulator circuit without either a balanced modulator or using phasing circuits to cancel the carrier frequency.

Lefty