[...] noise ratios are good, the residual readings will accurately reflect the separation.
However, you may find the residual signal easier to read if you bypass the pre-emphasis circuit in the generator.
When going this way, don't let yourself be deluded by apparently excellent separation at high frequencies. The decoder's de-emphasis always reduces the high-frequency output, but the generator is not increasing it to yield a flat signal. The dB value of the de-emphasis curve must be subtracted from your readings to yield the true separation.
Fig. 7--Tuner response curves, showing limits.
The output from the tuner should stay within the shaded portion of the curve shown in Fig. 7. The nominal values to subtract from your readings are listed in Table 1 as "dB down from ref." 'If you wish to measure separation more accurately, measure its response and use those values for the ones to be subtracted.
Connect the scope to one of the tuner audio outputs. Adjust the scope so that you can see the residual 19and 38-kHz signals. Adjust the 19and 38-kHz filters on the decoder for the channel you are watching to get the minimum output. Repeat for the other channel.
With the 19and 38-kHz filters in the tuner properly adjusted, the increase in the residual distortion from a mono signal to a stereo signal will not be more than a doubling of the percentage of distortion.
Example: 0.5 percent mono will not be more than 1 percent stereo. On high quality stereo tuners, you can expect little or no change.
If the internal 38-kHz oscillator is clean, and the i.f.'s are properly aligned, there will be no change in the distortion for a composite signal between the PHASE TEST, LEFT, and RIGHT signals.
As a final test, the tuner should perform just as well with a weak r.f. signal from the generator as with a strong one. Make the weak signal just strong enough to give a good signal-to-noise ratio. This is another area of difficulty, as an improperly aligned i.f. section will change its response with a change in the signal strength.
When the change is symmetrical, it is acceptable. Any other change is unacceptable for proper operation. This can be observed easily with the sweep generator.
See Fig. 6 again.
Another test is to check the separation and distortion at both 100 percent modulation and 10 or 20 dB down from 100 percent. There should be no appreciable change. If there is, something is wrong with the i.f. band pass.
R.f. Adjustment
Now comes the most difficult part. It is the alignment of the r.f. section. It is understood that some of the inexpensive tuners have their front ends so made that it is about impossible to make any adjustments other than one oscillator setting and one r.f. setting. If you are fortunate enough to have a more expensive unit to align, a complete front-end procedure can be performed.
The oscillator and the r.f. sections can be adjusted at the same time, thereby cutting down the time and effort required to achieve optimum results. First, the dial pointer should be located properly so as to cover the entire dial correctly. Second, the oscillator must be adjusted at both the high and low ends of the band for proper tracking. This is easily done by using either stations of known frequencies, or a crystal controlled or calibrated generator.
The oscillator trimmer is adjusted when tuning at the high end, and the coil is opened or closed when tuning the low end. The dial is set to the proper point corresponding to the frequency being checked. The oscillator adjustments are then made so that the tuner responds to the indicated frequency. To check the r.f. section(s), use the same signals as for the oscillator tracking.
Insert a tuning wand ( such as General Cement's tool #8278) through one of the r.f. tuning coils, or as close as you can get to it. Note the resultant meter swing.
If it goes down, try the other end of the tuning wand. This will detune the coil in the opposite direction. If the deflection is down by the same amount as previously, the coil is center tuned. Any other indication is one of mis-tuning.
The iron in one end of the tool increases the inductance of the coil into which it is inserted, and the brass end decreases it.
Closing the coil will increase its inductance, thus tuning it to a lower frequency, while opening it will decrease its inductance, tuning it to a higher frequency.
The tuning-wand test is used to check the r.f. tuning at both the high and low end of the band. As with the oscillator, the trimmers adjust the high end and opening or closing the coils, ( tuning them) adjusts the low end. So if you get an increased' meter indication, or AGC voltage, when inserting the brass end, open the coil to decrease its inductance (low-frequency end), or decrease the capacitance (high-frequency end). Conversely, if the iron end increases the reading, close the coil to increase its inductance, (l-f end) , or increase the capacitance (h-f end) . I have found only one tuner so far that used screw adjustments for the coils that prevented using the tuning wand.
The only solution was to note the screw slot position and then carefully rock the screw back and forth while looking for the maximum output. Several repeats will have to be performed at both the high and low ends of the band in order to achieve optimum results, i.e. no further improvement.
Having the cover on the r.f. housing is even more important for these adjustments than having the bottom plate on the tuner for the i.f. adjustments. Use the alignment holes provided. If none is available, remove the plate and drill them. Fortunately, a hole large enough to admit a tuning wand will not adversely affect the performance of the tuner.
On inexpensive units, you can sometimes improve the results by relocating the dial pointer and resetting the oscillator trimmer. You will have to keep track of the amount of AGC voltage you can get on some given station at the low end of the dial. The object is to try to move the high end around ( with the help of the adjustments, and resetting the dial pointer accordingly), so that a better low end might be achieved.
One note of warning. Don't try to improve upon the adjustments once you have achieved what is wanted. The inevitable result is that the attempt to improve soon causes a complete loss of whatever was obtained, and you must start over again searching for the desired setting of adjustments. This rule holds true for every section of the tuner, r.f., i.f., and stereo decoder.
Now you have the story on how to go about tuner alignment. You need only add a large measure of patience and your attempts will be successful, Good Luck!
----- TEST CHART
=============
(Audio magazine, Jul. 1970)
Also see:
How To Align Your Stereo Tuner (Part I)--Modifying the FM Generator (Jun. 1970)
How to evaluate FM stereo tuner performance (Jan. 1972)
"An Integrated Circuit FM Detector" (Feb. 1970)
Men of Hi-Fi--The Perfect FM Tuner (Jan. 1972)
New Specs for the New Tuners (Jan. 1973)
AM Stereo: An FCC Fiasco (July 1982)
FM Fidelity: Is The Promise Lost? (March 1985)
The Problem with FM (March 1985)
= = = =