Testing for Frequency Drift using ARGO Frequency Measuring Software                      

Go K0TO web site for Checking and Setting the Frequency of the FT1000MP REF oscillator and Temperature Stabilized FT1000MP REF oscillator which contains more information and graphics on this issue. (Updated May 4th, 2003)

A number of folks have asked about the ARGO frequency measuring software. Here is the URL to the site which has the downloadable program: http://www.weaksignals.com/argo You will need to connect the FT1000 AF out port to the LINE IN port on a sound card via a direct connection. No transformers are needed since you will be receiving not transmitting.To check your radio's frequency and stability do the following:

1. Manually enter 10.000.00 into VFO A. Be sure that you set the VFO so that frequency is exactly 10.000000 MHZ down to the single HZ. Read your manual
if you do not already know how to tune to the exact cycle rather than the 10 Hz value shown on the display. Select an antenna that allows you to hear WWV.

2. Set VFO B = VFO A.

3. Tune VFO A to 10.000.50 and select USB or LSB so that you hear the carrier tone of 500 Hz. (I am traveling and forgot if it was USB or LSB that will give the tone at 10.000.50 -- but one or the other will, probably LSB.)

4. Exchange VFO A and VFO B. Set VFO A to 09.999.50 on the dial and select the Mode (probably USB) that allows you to hear the 500Hz carrier. Now you
will have one VFO that is reading 10.000.50 and the other reading 09.999.50 and each has a different SSB mode (LSB/USB).

5. Switch back and forth between VFO A and VFO B. If your ref oscillator is set for the correct frequency the tone will be the SAME no matter which VFO
you are listening to. (If the tones are different, your reference oscillator is not on the desired reference frequency.)

6. Start up the ARGO program and using the help screens set up the parameters for your sound card. The Mode should be set at 3s dots and the Speed should be normal. Click the Start Button in the lower right hand corner and you will see a waterfall display. Look for a line at about 500 Hz and click on the line. The trace will move horizontally across the screen from right to left and the frequency will be shown at the top of the chart and at the right side. When you switch from VFO A to VFO B the frequency will change (unless you happen to be right on the reference frequency). Add the frequencies shown for VFO A and VFO B together and divide the sum by two. Subtract 500 from the result and you have a measure (in Hz) of how far off of the correct reference frequency your radio is. (Note: If your radio is 'right on' you will have added 500+500 to get 1000 and divided by 2 for a result of 500. When you subtract five hundred from that you get zero which
is the error in the ref frequency).

7. Turn off the radio and stop the ARGO display (button at bottom right). Let the radio cool for ten or 15 minutes. Turn on the ARGO display after cooling the radio and the turn on the radio. You should be able to see the reference frequency come back to the original value you measures. It may take some time. Take readings and write them down. Make a couple of runs this way to get a good idea of what the frequency error of your radio is. If you make a reading every minute you can get an idea of the drift in Hz per minute.

I will advise reflector readers when I have additional information on my web site (probably at least a couple of days). If anyone does these measurements and wants to share their results I would be interested simply to see what the variability between radios happens to be. You can send results directly to me or put them on the reflector == your call.

Tod, KØTO http://www.k0to.us/

I have posted two JPEG files which show the ARGO measurements for the first 90 sec after turning the radio on (cold start.jpg) and after 90 minutes of operation (long term.jpg). They can be found at http://www.k0to.us/

I have posted additional pictures of the ref oscillator mod that I made at www.k0to.us . The pictures have links in the table following the home page picture. The four pictures showing the reference oscillator board and the modifications are DSCN4434,DSCN4435, DSCN4442 and DSCN4443. The thermistor-resistor circuit is one that uses a 25V50/50C positive temperature coefficient thermistor, (Mouser part # 527-3006-25V50/50C about $2.28) in series with a 1/4 watt 68 ohm resistor connected between ground and the +9 volts on the board. One lead of the thermistor is soldered to the case of the crystal and from there to ground. The other lead of the thermistor goes to the 68 ohm resistor which has its other end connected to +9 volts. The thermistor should be physically touching the crystal case to assure good heat transfer. Remove the board before working on it. Scrape the metal case in a thin line so that you can put a bead of solder on the case (20 watt iron is plenty hot). Then solder the thermistor lead to the case along the solder line. Dress the ground lead to the bottom of the board per the picture. Connect the 68 ohm resistor to the +9 volt point shown on the picture. Use insulated tubing to protect the leads from shorting.

Supplemental Notes: A few days ago I downloaded the program ARGO (freeware) which allowed me to use my computer's soundcard to check the frequency accuracy and stability of my FT1000MP. I found that the MP reference oscillator was 50.6 Hz off after the MP had been warmed up for about 1 hour. I then used ARGO to make a snapshot of the drift in frequency from the time I turned on the MP until after the first half hour or so. My radio has the standard reference
oscillator board and does not have a special TXCO unit. The freq drift was about 3.5 Hz per minute reducing to about 2.5 Hz per minute 40 minutes after turning the unit on. The drift seemed to continue almost indefinitely.

I decided that I would see if I could reduce the drift and time needed to reach 'frequency stability' by incorporating a thermistor on the reference oscillator board. I fastened a thermistor obtained from Mouser electronics to the crystal case. The thermistor had a resistance of about 50 ohms when cold. To limit current drawn from the 9 volt source going to the ref osc I put a 68 ohm resistor in series with the thermistor. The current drain is limited to about 75 ma at start up and drops to about 40 ma when the thermistor warms up.

The result was that that within 90 seconds the ref osc frequency now is within 2 Hz of the final (desired) value. After five hours the frequency was still within 2 Hz of the desired value. The ref osc operates at about 10 MHz. A frequency that is within 2 Hz of a desired value of about 10 MHz has an accuracy of 0.2 ppm. The most expensive of the Yaseu TXCO's has a 0.5 ppm specification.

It may be that I am the only one who has an interest in this. However, just in case I am posting my results to the reflector. If anyone is interested I can put additional 'how I did it' information on my web site.

Tod Olson, KØTO http://www.k0to.us/

My apologies to those on the reflector who might have tested the ARGO program as Gary seems to have done. I happen to be in Minnesota for the next several weeks and my FT1000MP is back in Idaho. That accounts for me forgetting whether something should be USB or LSB as well as having forgotten that prior to doing the frequency measurements I went into the menus and set the non-transmit offsets to zero so that there would be no chance that some tweaking of the frequencies done by the CPU would produce errors in my readings. One of the Menu settings I changed was in group 8 and the other was in Group 9. In both cases I wrote down the values that were originally set in the menu. The changes I made were to set the RX-LSB and RX-USB offsets to zero. After the measuring and the modification to the ref osc board I restored the original offset values.

Gary comments that he can hear a difference in 'tone' between LSB and USB. My note on doing the measuring indicated that unless the ref oscillator frequency is set exactly where it should be there will be two different frequencies measured by ARGO. One of these will be less than 500 Hz and the other greater than 500 Hz. Most folks will be able to hear this difference very easily. There is a trimmer on the ref oscillator board (TC4701) and if you adjust the trimmer (very small amounts please) you will hear the tones get further apart or closer together. This is one way to move the reference oscillator frequency very close to the desired value even though you do not have a high precision counter. If you look at the ARGO values as you tune the trimmer you should be able to see and hear when you are within 2 Hz of the 500 Hz values for the LSB and USB readings that confirm the oscillator is set to the correct frequency. If the radio has not warmed up sufficiently
there will be drift off that point. In the case of my radio, the initial frequencies were 410 Hz and 690 Hz. This gave me an estimate of 50 HZ off frequency (410+690=1100. 1100/2 =550. 550-500 = 50Hz). As I tuned the trimmer to the correct frequency the frequencies became symmetrically positioned above and below 500 Hz and converged on 500 Hz. Once I had adjusted the trimmer to get the correct ref osc frequency I was able to look at the change of freq as a function of time following power on. The initial power on frequencies were more than 75 Hz from the desired value, but they were symmetrical in that one was 575Hz and the other 425Hz and they slowly converged toward each being 500 Hz. The rate of change seemed to be about 3 Hz per minute slowing to 2 Hz and then 1.5 Hz per minute. After 45 minutes the radio was still drifting slowly toward the desired frequency as the ref oscillator crystal slowly warmed up.

For those folks who wish to tweak the ref oscillator trimmer, you will need to remove both the top and bottom covers of the FT1000MP. On the right hand side (as viewed from the front panel) at the very front of the side mounted boards, is the reference oscillator and the trimmer, TC4701. A picture of the reference oscillator board in its normal position in the radio is at www.k0to.us . Go to the JPG links following the home page picture and select DSCN4434.jpg . That picture will show the trimmer at the bottom of the board next to the crystal .

Tod, KØTO http://www.k0to.us/

This site was last updated 08/21/07