[R-390] Crystal check
Jim Whartenby
old_radio at aol.com
Wed Jan 8 12:04:45 EST 2025
John
Spec is spec. It seems that all of the frequency instability in the R-390 is in the PTO. Over the full temperature range of -40F to + 149F the PTO should drift less than +/- 300 cycles. At a stable temperature, the frequency drift is perhaps +/- 10 cycles. The frequency instability of the PTO is a factor of 10X worse then the crystal controlled oscillators. A decent synthesized radio has perhaps less then one cycle of drift. The KWT-6 family is an example of an early synthesized radio. It takes some 40 tubes and even more transistors to get that one cycle of stability in a 400 pound, seven foot high rack!
I guess that if you notice a frequency drift then there is a problem. I doubt that I would notice a 300 cycle drift over a 24 hour period when the IF bandwidth is 8kc and listing to music. As in intercept receiver for voice or music, the R-390 is great. For a set and forget receiver for crypt traffic, not so much.
As time goes by, the weight of the R-390 becomes more of a concern for me but the same can be said for the R-1051! I have passed on all of my R-1051 receivers but have kept the SC-900 series examples since they are the earliest version of the design.
Regards, Jim
Logic: Method used to arrive at the wrong conclusion, with confidence. Murphy
On Wednesday, January 8, 2025 at 08:04:05 AM CST, jkharvie <jkharvie at verizon.net> wrote:
Thanks for your input Jim,The R1051 is a nice receiver, but I unloaded the one I had as I enjoyed the smooth frequency adjustment of the R390 much better, less wrist stress. Clearly all the digital electronics in the R1051 receiver yielded additional frequency accuracy and stability.Thanks for the link to the CR-36 A/U , helpful.Any thoughts on what was the max limit that crystals could be from design set-point before the receiver operation would be out of acceptable depot level servicing? The 2kHz variance did not appear to have much effect on ability to tune and operate the unit. I did swap out the 15,000 kHz crystal as this variance was 10x worse. My quick look assessment was the first time I have had a digital frequency synthesis with control, resolution and accuracy suitable to see how these crystals performed as a function of precise frequency adjustment, this was very educational and quite fun.Thanks againJohn N3JKE
Sent from AOL on Android
On Tue, Jan 7, 2025 at 2:52, Jim Whartenby via R-390<r-390 at mailman.qth.net> wrote: Bob is spot on, the crystals used in the R-390 and R-390A are CR-36A/U per https://www.r-390a.net/r390a-cr.pdf, there are three frequency tolerances given:
One is for when the ovens are used, +/- 20 cycles per megacycle over the temperature range of 80C to 90C.
When the ovens are not used, +/- 80 cycles per megacycle for the temperature range of -55C to +80C.
Then there is the normal frequency drift over time, +/- 5 cycles per megacycle.
I would think that operating at a normal room temperature of 25C for a few hours, one could expect that the crystal oscillators would settle down to better then +/- 25 cycles per megacycle which would be the combination of the first and third frequency tolerances. This would be good enough for SSB and radio teletype but not good enough for encrypted communications where 1 cycle per megacycle or better frequency accuracy would be needed. This need for frequency accuracy was easily met using the R-1051.Jim
Logic: Method used to arrive at the wrong conclusion, with confidence. Murphy
On Monday, January 6, 2025 at 12:20:47 PM CST, Bob Camp <kb8tq at n1k.org> wrote:
Hi
The resonant frequency you “see” in a test set is a result of the crystal *and* the load capacitance of the test fixture. More or less: The capacitance of the fixture “tunes” the crystal.
Next up: If you have net C on both the input and output side of the fixture and are looking for peak amplitude, that also will impact what you see. You have built a “pi” network and it will match up best at some very specific frequency.
0.00015% is 1.5 ppm. When manufactured, these crystals likely had a tolerance of +/- 20 ppm at the reference temperature.
Depending on what crystal you are looking at, the reference migh be temperature like 85C. The room temperature number would be (many) 10’s of ppm off from there.
The temperature in the radio will be warmer than room temperature. With the ovens off, it will be cooler than 85 (or whatever) degrees. The same “many 10’s of ppm” gets into the act here. What you see in the radio will not be what you see on a bench at 25C.
Further complicating this is that they got tested in some very specific military test sets. Just what load those test sets presented when calibrated “by the book” was very much a matter of great debate. The answer back in the day was large banks of “reference crystals” that got used to keep everything on the production line reporting the same numbers. Replacing those test sets was not at all easy …
Quick ppm math: 1 ppm is 1 Hz at 1 MHz. If you have a 10 MHz crystal and are shooting for 1 ppm, you look for 10 Hz. Way way easier than playing with percent.
Fun !!!
Bob
> On Jan 6, 2025, at 12:50 PM, jkharvie via R-390 <r-390 at mailman.qth.net> wrote:
>
> Making progress with servicing my EAC receiver and I thought it might be of some value to know how the crystals in the deck have aged regarding frequency and amplitude performance.
>
> I am not aware of any specific language regarding how far out of spec the crystals can be and the receiver still operating well but this is what I found.
>
> Using a scope and digital signal generator I conducted an evaluation of each crystal with a quick assessment
>
> Results as follows
> Procedure for Crystal frequency performance assessment
> Tools needed
> Digital Signal Generator (0-60Mhz, 2 channel, multi waveform, 0.01 Hz resolution)
> Oscilloscope - Tektronix TAS 475 (100 MHz- 4 Channel)
>
> Procedure:
> 1) With care remove crystals from receiver, note placement position and orientation. Label each crystal for convenience
> 2) Set signal generator to 0.4 volts, waveform to sine wave
> 3) Hook scope up to crystal in series with signal generator
> 4) Adjust scope for best waveform presentation
> 5) At the precise frequency of resonance, the measured signal amplitude will increase, thus adjust signal generator frequency for highest crystal resonance peak
> 6) Record frequency of crystal when peak is measured
> 7) Replace crystals into receiver
> 8) If error is higher than 0.00015% offset from design frequency consider replacement of that crystal9) Note the voltage, at resonance the peak should be clearly higher
>
>
> Crystal Number)Labeled Frequency(kHz)_resonancefrequency (kHz)_"delta(kHz)P"_"delta(kHz)F"
> 1)9000.00_9,000.78_0.78P (pass)
> 2)9500.00_9501.22_1.22P
> 3)10000.00_10001.2_1.20P
> 4)10500.00_10501.52_1.52P
> 5)11000.00_11001.72_1.72P
> 6)11500.00_11501.21_1.21P
> 7)12000.00_12001.82_1.82P
> 8)12500.00_12501.44_1.44P
> 9)13000.00_13001.74_1.74P
> 10)14000.00_14001.96_1.96P
> 11)14500.00_14501.82_1.82P
> 12.1)15000.00_15021.83_X_21.83F (fail)
> 12.2)15000.00_15000.07_0.07P (Replacement unit)
> 13)15500.00_15502.11_2.11P
> 14)16000.00_16002.15_2.15P
> 15)17000.00_17001.94_1.94P
>
> Thoughts?
> AlsoHappy New Year to everybody!
> ______________________________________________________________
> R-390 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/r-390
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:R-390 at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:R-390 at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
R-390 mailing list
Home: http://mailman.qth.net/mailman/listinfo/r-390
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:R-390 at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
More information about the R-390
mailing list