Crude frequency alignment of SSB using only a second radio....

[144inBama]

Well, looks like I get to play around with some other stuff tomorrow. I do know that I can put one on AM and the other on LSB and it's totally coherent both ways. LSB to AM does have a deeper sound but you could totally hold a conversation. Both radios, the untouched being a Uniden 980, transmit on the same freq when using a tone.

 

[Handy Andy]

Now that you have a radio in which to hear what you sound like on SSB modes, you also know their term they may use on the air, as you sound like you're on AM - is because you're close (soo close) in frequency that they don't hear you sounding like Billy-goat gruff or Jacques Cousteau on Heli-Ox...

​

 

[144inBama]

Now that you have a radio in which to hear what you sound like on SSB modes, you also know their term they may use on the air, as you sound like you're on AM - is because you're close (soo close) in frequency that they don't hear you sounding like Billy-goat gruff or Jacques Cousteau on Heli-Ox...

View attachment 42313​

Is it normal-ish for SSB to be off frequency when 1st turning on the radio? It's been off by .006 until she warms up. Say 27.204 (20 LSB) will be 27.2046 which sounds bad on the receiving radio but if I just let her sit idle for about 6 minutes or talk on AM for a couple of minutes, she'll level right out and hold steady at 27.204. Is there anything I can do to add stabilization to it or is this just the nature of the old 148s?
Oh, and one other thing....
I've seen so many different bias values it's confusing. I've seen 25/50 35/50 35/60 40/100 (not gonna try that one) and your suggestion of 50/50 (which it's currently set at)...can you, in "I'm talking to an idiot" terms, explain how these values affect the performance of the radio?

 

[Handy Andy]

Just to help, Yes, Xtals and their oscillator circuits they control - are funky things.

On top of and being the older chassis that uses the older designs - so the 148 or 2000, or the Grants or really any AM or AM/SSB radio - even Galaxy - exhibit this trait. It's a environment thing;, temperature, humidity and time - all work against the oscillators trying to run - they do warm up, and that's why they warm up, vibration - so as they ring, vibrate of oscillate - that energy is also converted to heat, which then helps (or hinders) the vibration until it becomes stable, able to release the energy it gains from the vibration - en equilibrium - stability.

The hindrance is from the fact that it's harder to vibrate when cold, than when warm - think of this like Water, in normal pressure air, ice freezes at 32 F - but boils around 212 (Sea level) - in between, it's liquid - not solid.

So to make the analogy, ICE is the solid of Water, but to "break" the Ice, it needs to warm up to reach an ability to vibrate easily and with less energy than when it's cold.

• I used to suggest a simple light bulb, since they are filament and heat up, I've used them in Grants and even Emperor TS-3010's as a means to help warm up the PLL/VCO section and to "thaw out" the LCD display (another good analogy to ICE versus Liquid LCD's displays improve with warm air versus cold - they're sluggish when cold)
• I got laughed at for this effort, but the effort is used elsewhere- they call it "ovens" that are for the EXACT timing references - heat (an oven) is used to make the oscillator stable at a high temperature - higher then usual, not by much though. It is for the reason of less change in the vibration at that temperature - they make the system - it's feedback inputs, the timing references and the output - all are controlled using the feedback principles to control the frequency.
• The light bulb simply mounted to a switch I didn't need or use (like Tone) and since it was switchable, made it easier, faster to warm up the radio in a cold vehicle - it took more power to do so, but that heat the bulb released helped the radio stabilize faster and quicker for longer QSOs' on cold nights when you're working 2nd or 3rd shifts and it's winter - you only have so much time before breaks are over and its' back to work.
• The heat the light bulb gave off, helped me have QSO's while on breaks. Plus as you will learn later - the QSO timeframes for contacts can be short, so if you're in the right place at the right time, being on frequency helps you establish that contact or get a radio that would otherwise be unstable for wintertime use in the mobile; work in that environment, so you can enjoy the radio and it's features when conditions aren't favorable.

As far as Bias? The effort many used to make (their Finals in their equipment) is to push a lot more power out thinking they'd be a lifelong supply if they burn out, nowadays, you have to trim them back more towards their original designs and that also improves the lifespan, or Meat Time Between Failures - MTBF - they simply last longer.

If you push them to their limits, they may produce more power- but as ghost watts, minor increases than true power they're supposed to generate - you can run into clipping problems easier when the Bias is too high. An example would be - like a mic set too high, it's overmod and it's harder to listen to and understand. They will produce a lot of heat and wind up with a shorter lifespan and then the OEM you like, is not what you once had,...you have to fix it - it's also obvious that they can fail when you need them the most.

Turning down the bias to their mid-range is about the best...as least for the life of the radio.

 

[Shadetree Mechanic]

I use this method using two radios with dummy loads and my NESDR Smart usb dongle in my laptop computer. One khz tone on my phone. Tune xmit on first radio to sdr then rx on second radio to tx on first. Then tx on second to sdr then rx on first.

I do have an HP 5315A counter, I just need a good way to verify it's accuracy.

 

[secret squirrel]

May need NIST certified reference radio like stated before.

 

[whiteastro]

This thread is talking about frequency adjustment on 38 LSB so I have a question. I have a President McKinley that was never right out of the box and the factory sent it out for frequency adjustment. The radio is much better but still needs a little help to be clear on 38 LSB. I have a Dosy 3001 S meter that says the frequency reading is 27.3854 when I key the mic instead 27.3850 like it should be. Is that enough to make it sound off frequency and if so which pot inside the radio can I slightly tweak to put it right on. I know if I use that meter on my other radios they read right on 27.3850. I hate to send the radio out for adjustment again if it just needs a feather tweak. LOL Thanks

 

[Handy Andy]

I could get Fired for this...

But it's the other guy you worry about...

IN every case for this or any SSB radio - you'll need to use a simple technique to tweak - using a simple eyeglass screwdriver.


Which McKinley you got? EU or USA?

​

Because what I'm trying to help you with, is subjective, because if you're not comfortable with doing this check yourself - get rid of the radio and use one with a direct access tunable VCO for this instead. Don't want you to be shown a technique that gets you in over your head and expen$ive repair due to the mal adjustment that needs re-adjustment and re-alignment

You're saying it's a little off, when you have to find the coil that makes this a little off...

Being the safest way, is to simply take the metal tip of that small eyeglass tool and find the coil - in TX mode - that CHANGES the tuning - which way up or down - doesn't matter - just find that coil that changes the tuning when you put that tool tip in - don't turn the slug - just try to find the coil that is causing this change you don't like. The one that changes it's frequency due to the change in the inductance (additional metal the tool tip provides) is the one that AFFECTS the mode you're in. That is the one you'll need to mark.

So to help...

• set up the system
  • Set your channel on the RECEIVER side to the 38 LSB
  • Set the McKinley to same channel same mode

Key up your Mckinely - into a dummy load for safety - and simply hold the mic and keyup and apply some simple "ambient" audio - music from a tabletop radio or whatever and just - gently - insert the tool tip into those three slugs and listen - not to tune, but to LISTEN for the change in tune thru that other RECEIVER - that change is due to the proximity of the tip into the tuning coils in the strip used for the IF section to SEND the signal out - it needs an IF signal to even start the mixing process - but you have to find the one to tune or at least SHOW them which one to "tweak".

Now, if you have the STOMACH for this, you can then use the Receive radio for the REFERECES in both AM mode - and LSB mode for the McKinley.

Only thing? You tune the McKinley - in AM Mode - but leave the RECEIVER radio in the LSB mode....

Set up like above, but the McKinley in now in AM mode - listen to that audio in the RECEIVER - ADJUST RECEVIER's Clarifier to "beat" (best sound) for the McKinley AM mode to hear the LSB side of the McKinley's AM side for the best most clarity in the RECEIVER.

Unkey and put McKinley unto LSB mode again...

Repeat key up audio and knowing which slug, you can use a ceramic tip tool to alter - tweak that coil to "beat" (Best sound) like you did for AM side only you're leaving the McKinley alone and just tweaking the IF section to bring the audio in line or more resonate.

This is from a PC-122 - a radio from a PREVIOUS
generation that used PLL-IF and AF-IF to mix together
​

• In the above photo - the older radios used pairings One IF from the PLL one AF -IF from the Balanced modulator. But both "pairs" of coils worked for only one mode and only one did the Audio the other did the PLL IF​
• You work with the Audio - leave the PLL alone​

It's that simple - but to do it right you need to know which coil - by identifying it thru the metal tool tip test trick - then tweaking that coil to re-align the mode for best sound thru the RECEIVER.

 

[whiteastro]

Which McKinley you got? EU or USA?

I have the USA model ! I'll have to study before I jump it both feet ! LOL The radio is Right on frequency in AM mode !

 

[Lkaskel]

Bama,
One thing that I did not see get addressed here is the actual frequency you are asking about and why a 6 digit freq counter can only help to a point. When you ask about 38lsb and reference 27.385 (or 27.384) the actual complete frequency is 27.385,000. I struggled with this when I began trying to figure out how to align a radio. In many manuals they will ask you to align your transmit to 27.385 +/- 20Hz. That means they are asking for you to adjust from 27.385020 down to 27.384980 or for lsb 27.384020 down to 27.383980. The reason I bring this up is you are asking about 27.2046 as an example from above. The actual amount it is off frequency is 400Hz if you are expecting 27.2050. It looks like this in reality, 27.204600 instead of 27.205000. I hope that makes sense. I needed to write it down when I began because I was used to having freq counters on radios and thought they were telling the whole story. You really need an 8 digit or more counter so you can see what is happening down to the 1Hz level.

On SSB, if you get off frequency 10-20Hz it starts to sound bad. One way to test this out is to find an SDR on the interweb that you can listen to. You can try this site as an example http://kiwisdr.com/public/. Make sure that you are looking for something on side band. Find a conversation and adjust the frequency on the SDR so it sounds perfect. Now start to adjust the frequency you are listening to by 10Hz at a time. It really brings it into clarity how minor frequency changes affects what others hear.

For this instance I like the idea of having the clarifier track with receive (known as unlocking the clarifier). That way you can align the radio for receive and it should be fine on transmit. Just adjust your clarifier to be on frequency and your transmit will be the same.

 

[Handy Andy]

@Lkaskel and @144inBama - one, or two actually (maybe three) - thoughts on the issue of ACTUAL working frequency, versus the intended - specified tuned frequency.

IF you're in a hurry, or just don't want to deal with a headache, skip this message - else I'm leaving this as a LEGACY so others that happen to wander in can find - read and hopefully derive some information as not just "How" radio works the way it does - but to help understand the "Why" is has to be done in ways that add, or subtract errors in the processing.

There are aspects, or viewpoints, to the observer in this.

You have;

• the audio signal processing in the Transmitting radio
  • - going out at as a range of frequencies (not specific when you're dealing with the intended bandwidth or range of frequencies) specific to a channel (audio shifts in the region - again not specific)
    • - by using one frequency to mix the audio into an IF the transmitter is able to use to tune its own PLL (based) IF
      • - using two frequencies mixed upwards into another one
      • - being close to the center of the channel
      • - but not exactly at the center of that channel.

​

• The Receiver - listening in to a range of frequencies at the channel - but only able to hear an expected range of frequencies within that bandwidth of channel - using one or TWO - frequencies in which to decode the information - by applying a signal from the LOCAL Oscillator - can remove the IF and extract the Audio Information as DECONVERSION which is that audio from the RF signal.

​

• The Frequency counter as a source of the measurement of the frequency the Receiver (or Transmitter) is using to mix one signal (audio) into another (IF) and then onto another (PLL IF) with its mixed result upwards to the bandwidth of frequency of channel.
  • Being that if the Counter uses its own Crystal as its BASE reference and then calculates using the IF-frequency (which is lower than the expected - so it is a prediction) what the new up conversion frequency is using only IF as is sample - a ppm ± measurement error issue.

So, in a basic sense - you have 3 sections of signal that introduces a level or error in processing - some call this DRIFT while another approach is from the error in SAMPLING (Alias - or Aliasing).

Once you see where the processing of Audio is the actual problem - you can set up the IF and all the filtering and even the PLL offset (10.925 - 10.965 - 10.700) to obtain the image for Transmitter and even the receive - but the ERROR itself is in where you CENTER your audio signal into the IF - due to predominate perception of tone - it's not an error in the processing - it's an error in the admittance/acceptance and tuning of it.

To blame the tune-up on the tech is really not fair to those whom have little information to go on, so the tech themselves must take into account the desired results and tweak the mixing products properly to make sure the audio can be derived and heard "normally"

 

[kopcicle]

Bama,
One thing that I did not see get addressed here is the actual frequency you are asking about and why a 6 digit freq counter can only help to a point. When you ask about 38lsb and reference 27.385 (or 27.384) the actual complete frequency is 27.385,000. I struggled with this when I began trying to figure out how to align a radio. In many manuals they will ask you to align your transmit to 27.385 +/- 20Hz. That means they are asking for you to adjust from 27.385020 down to 27.384980 or for lsb 27.384020 down to 27.383980. The reason I bring this up is you are asking about 27.2046 as an example from above. The actual amount it is off frequency is 400Hz if you are expecting 27.2050. It looks like this in reality, 27.204600 instead of 27.205000. I hope that makes sense. I needed to write it down when I began because I was used to having freq counters on radios and thought they were telling the whole story. You really need an 8 digit or more counter so you can see what is happening down to the 1Hz level.

On SSB, if you get off frequency 10-20Hz it starts to sound bad. One way to test this out is to find an SDR on the interweb that you can listen to. You can try this site as an example http://kiwisdr.com/public/. Make sure that you are looking for something on side band. Find a conversation and adjust the frequency on the SDR so it sounds perfect. Now start to adjust the frequency you are listening to by 10Hz at a time. It really brings it into clarity how minor frequency changes affects what others hear.

For this instance I like the idea of having the clarifier track with receive (known as unlocking the clarifier). That way you can align the radio for receive and it should be fine on transmit. Just adjust your clarifier to be on frequency and your transmit will be the same.

Some years ago WB8JKR, Mark Graalman (SK), pretty well covered it . For sure the master frequency adjust should zero beat some reference. However that isn't all that needs to be done.
Getting upper and lower sideband carrier insertion points located properly with respect to the filter passband can be a real mindfuk for the uninitiated.
The first mistake the uninitiated make is to get out the book and do it by the numbers and wonder why it didn't work.
The second mistake is leaving all kinds of wires and probes laying in the radio
Lastly the probe itself loading the oscillator circuit.

I've rewritten this for specific radios with 7.8~10.7 filters but this time I'll just leave it here the way Mark wrote it for the Heathkit HW-101 .



I.F. Filter Passband Improvements:
I believe the crystals that Heath supplied for the carrier oscillator were of fairly wide
tolerance, thus the frequencies of the LSB/USB/CW carrier injection may not be properly
positioned on the slope of the I.F. filter. This can affect both the receive and transmit
audio response to a great degree. Telltale signs of this are not having the same audio
response on USB and LSB, and reduced CW power output when using the CW filter.
It is relatively easy to tell if the USB/LSB carrier insertion points aren’t placed
equidistant from the center of the I.F. filter passband. After the rig reaches a stable
operating temperature (1/2 hour) disconnect any antenna and peak the preselector for
maximum receiver gain. Next turn up the volume control to a slightly higher than normal
level and listen closely to the hiss coming from the speaker. Then switch to the opposite
sideband. The pitch of the receiver background noise should be the same if the USB &
LSB carriers are both placed equidistant from the filter center frequency.

8
If the carrier oscillator frequency is placed too far from the filter passband, the receive
and transmit signals will lack “lows” but the opposite sideband rejection will be high. If
the carrier oscillator frequency is placed too close to the filter center frequency, the
receive and transmit signals will have excessive “lows” and the opposite sideband
rejection and carrier suppression will suffer. Balance is the key.
On my particular HW-101, the actual measured carrier oscillator frequencies were 3393.8
kHz LSB, 3395.9 kHz USB, and 3395.17 kHz for CW. This resulted in a “tinny”
sounding audio response in LSB compared to USB, and a very “bassy” sounding USB.
The CW power output while using the SSB filter was 110 watts, but since the CW carrier
oscillator injection was so far from the CW filter center-frequency of 3395.4 kHz, the
CW power output was 50 watts while using the CW filter!
Heath’s intended frequencies for the carrier oscillator were 3393.6 kHz LSB, 3396.6 kHz
USB, and 3395.4 kHz for CW. With the specified filter center-frequency of 3395.0 kHz
the USB & LSB carrier positions would be 1.6 kHz each side of the filter center-
frequency. Unfortunately, the filter center-frequency may not be exactly 3395.0 kHz, so
simply placing the carrier injection points equally-spaced from 3395 kHz may not have
the intended result. In order to determine the filter center-frequency one must balance the
audio response between both sidebands, measure the USB and LSB carrier frequencies,
and finally subtract ½ the difference between the USB & LSB frequencies from the USB
carrier frequency. The result will be the I.F. filter center-frequency as it exists within
your particular rig. For example, if the audio response is exactly the same between
USB/LSB, and the measured USB carrier frequency is 3396.31 kHz while the LSB
carrier frequency measures 3393.51 kHz, then the difference is 3396.31 - 3393.51 = 2.8
kHz. One-half the difference is 2.8 ÷ 2, or 1.4 kHz. Then the USB frequency of 3396.31
kHz - 1.4 kHz = 3394.91 kHz, which in this example is the actual SSB filter center
frequency.
I like audio with a tad bit more bottom end response, so I placed my carrier points just a
little closer to the filter center frequency than the Heath spec., i.e., rather than 1.6 kHz off
center I went with 1.4 kHz. The frequency of the oscillator is lowered by placing a small
amount of capacitance in parallel with the crystal, and the frequency is raised by putting
capacitance in series with the crystal. To put a capacitor in series with the crystal simply
cut one circuit board trace just before the crystal pin as indicated, and solder the capacitor
across the opened trace (Figure 4). A 100 pf capacitor in series will move the crystal
frequency up about 100 Hz, but the same frequency change in the downward direction
would only require about 10 pf connected in parallel with the crystal.
On my HW-101 I put the capacitors (silver mica’s) directly on the circuit board foils. I
got one sideband to sound the way I liked, and then simply adjusted the other sideband to
match it in audio response. I wound up using a 10 pf cap in parallel with the LSB crystal,
100 pf in series with the USB crystal, and 80 pf in series with the CW carrier crystal.
Following these changes, the new carrier oscillator frequencies for my rig are 3393.51
kHz LSB, 3396.31 kHz USB, and 3395.38 for CW. The audio is perfectly balanced
when switching between sidebands, indicating a true I.F. filter center frequency of

9
3394.91 kHz. The CW power output while using the CW filter went from 50 watts to
110 watts. I also soldered a short loop of wire to the center lug of the carrier null pot to
serve as a test point to measure the carrier oscillator frequency. Be sure that once you
have determined the filter center-frequency, you place the oscillator frequencies no closer
than about 1.4 kHz and no further than 1.6 kHz from the filter center-frequency.
To recap, we are actually matching the response between USB & LSB by ear, then
verifying with a frequency counter that the carriers are no closer to the filter passband
than 1.4 kHz, and no further than 1.6 kHz. Even though we can match the pitch between
USB and LSB with our ear, we can't tell exactly WHERE they are -- only that they are at
the same point on the filter slope.

~WB8JKR, Mark Graalman (SK)

So what Mark was trying to tell us is that most of this can and should be done by ear. Call it crude and off script but there you have it.

One last bit of anecdotal humor. I had put hours into restoring a classic base station for a friend I had introduced to CB/SSB. I spent the time getting this radio to talk where it listened to within a few ppm.
The first few times I talked to him it appeared that the radio (or mine) was noticeably drifting. Every few minutes I'd have to touch up the clarifier to keep up. This discussion led to "you're drifting , no you're drifting, NO YOU'RE Drifting!"
So that week end I went over to his house so I could talk to my base and see (hear) for myself. I quickly tuned my base in and said hello to my sister. Absolutely FM quality. I got up to refill a coffee cup an the first thing he does when he sits down is readjust the clarifier?!
"Why did you readjust the clarifier??!!"
"Because it didn't sound like sideband!!"
It took myself and a few other operators to cure him but in the end it all worked out.

Okay one last tip.
Zero beating to another radio can be a bit of a biotch if the pitch of the received signal isn't the same on USB as it is on LSB. That's the whole reason for the post .

 

[kopcicle]

To blame the tune-up on the tech is really not fair to those whom have little information to go on, so the tech themselves must take into account the desired results and tweak the mixing products properly to make sure the audio can be derived and heard "normally"

A good tech will get out the tools and put the radio on the numbers.
An artisan will identify the filter passband and adapt the carrier insertian points to match.
Neither one is wrong but we know which receiver sounds better
Tnx Andy

 

[Xjeepguy]

The title is my question. Is this possible, at least in theory? While I do understand the extreme crudeness of this idea, I can't help but think it's possible with somewhat decent (not perfect) results. With the clarifier on both radios in the center position, using the known good radio and an external 1K tone through the mic transmit and adjust 2nd radio RX pots until it is clear on the receiving radio. Then transmit the same 1K tone on the 2nd radio and adjust 2nd radio TX until it is clear on the known good radio. I'm asking this as a no other option, no techs, no equipment, kinda question.

Thanks for any input.

I've done exactly this, quite a few times. All with wonderful success. It works well for those radios that are just a TINY bit off. It's not a recommended method by a long shot but does work. I'm trying to locate the right xtal trimmers in my old 2950 to do just this. I think I know ( lol ) but was asking here in another post to verify I have the right trimmers to adjust. It's a different layout than the CB's I've done this to. Be careful and good luck.

 

[Xjeepguy]

When dealing with RF in this manner, how significant is ±50Hz? Would it be noticeable to all, most, few, a couple?

It depends on you and your hearing IMO. It drives me nuts just a hair off frequency. But others ( if you've ever listened to channel 38 LSB ) don't seem to have the same fussiness haha. Man, there are some bad bad off frequency examples on that channel.