|CRYSTAL GRINDING WITHOUT TEARS
Helpful Hints on Bringing Blanks to Frequency
Many hams have asked me for advice on how to grind old surplus crystals to new frequencies. They assumed I was some form of an expert on the subject just because I was the son of the owner of a crystal manufacturer. I spent part of one summer during high school working in the calibration department at Bliley Electric. I spent weeks calibrating the SMC100 dual frequency crystals blanks by hand. My production rates were horrible but I managed to finish far more than I broke. That is as much experience I have ever had in grinding crystals. I learned it is a laborious process that is part craft and part luck.
I recently came across this article in an old QST magazine in the personal papers of John M. Wolfskill, W3QKT, Bliley's chief engineer. I believe, John saved it for nostalgic reasons--a process he once practiced in his personal and professional lives. I hope you find this article a simple and realistic guide that is helpful and encouraging.
Chuck Bliley, K3NAU
Whether or not You acquire some government surplus crystals, there usually comes a time when you want to move a crystal to what seems to be a greener spot in the band. You'll get there with fewer tribulations if you have some idea of what crystal grinding is all about. Here Is some good dope from a ham who is associated with a well-known crystal producer.
Before making crystals professionally, the author's feelings about crystals were probably the same as those of most hams. Eventually daring to open some of my commercial crystals, I decided to grind one of them to a higher frequency. After a terrific struggle the frequency was finally raised - so high that the crystal was never used again!
In the past three years I have learned enough about grinding crystals to know that it is possible to get into lots of trouble grinding your own. The only way to be a crystal maker is to work at it. However, it is hoped that the following information will help the average ham to grind his own crystals, although it does not guarantee to show you how to get all your blanks on frequency with good activity. If anyone reading this does know how to do this I would like to hear from him.
Fig. 1--A crystal dimensional chart. showing frequencies at which harmonies of low-frequency vibration modes occur as a function of the dimensions of the crystal blank. X and Z harmonics are shown by solid and dashed lines, respectively. The crystal dimensions should be chosen so that the length and width do not fall on one of the lines at the desired frequency, because coupling between the various modes causes erratic operation with temperature changes and is accompanied by a drop in activity. This chart is given for illustrative purposes only, since the dimensions at which couplings occur will be modified by slight variations in the angle of cut, parallelism of faces, whether or not the edges are beveled, etc. (Adapted from a chart prepared by the Franklin Engineering Co.)
Page 58 of the 1945 Handbook gives. an elementary idea about what crystals are and how they work. Most ham crystals are now either AT or BT cuts, and Handbook Fig. 262-A shows just how these two cuts oscillate. However, there are other vibration modes which can couple to the shear mode and cause interferences which prevent oscillation. The dimensional chart, Fig. 1, shows how these interfering modes can be avoided by using proper dimensions. Since these charts are based on computations it is usually necessary to check them by dimensioning and temperature testing a few crystals to determine the extent to which practice deviates from theory. The regions on the chart which are shown to be clear of interfering modes are usually explored before cutting blanks because in mass production the only way to get a practical percentage through the temperature test is to predimension all crystals. For practical purposes, however, usable activity at room temperature is all that is required in an amateur transmitter because temperature changes are of small magnitude.
A good crystal-grinding layout for hams should have several components. The first necessity, of course, is a flat piece of plate glass, usually about four inches square. In order to keep the crystal flat a "button," also of plate glass, is necessary; it may be either round or square and should be slightly larger than the crystal blank, as shown in the photograph. Both plate and button can be obtained at the local glass store. Two grades of abrasive, No. 303 emery for surface grinding and No. 600 Carborundum for edge grinding and beveling, are ideal. These can usually be obtained at a hardware store or at an opticians' supply house. A small paint brush is handy for moistening the abrasive and spreading it around the lapping plate. For those hams who have a micrometer and like to use it, the formula on Page 58 of the 1945 Handbook tells how to find the frequency when the thickness and constant are known. (The constant for BT quartz is 101.5.) A micrometer isn't really a "must" in grinding crystals; if you know the approximate frequency of your blank, all you need is your crystal oscillator and a receiver. The receiver, of course, should cover the frequency at which you will be working.
Because frequent checking of activity and frequency axe necessary while grinding the crystal, it is wise to provide a test holder and clip to make this process as rapid and easy as possible. A simple one made from an FT243 holder is shown in the photographs s; it is made by substituting a sliding cover for the screw-on cover plate. The sliding cover is nothing more than, a rectangular piece of sheet copper or brass bent to fit around the holder as shown. Crystals can be, interchanged, with this cover, in a matter of a, few seconds. For other type holders similar clips can be easily made from an old piece of aluminum. or sheet tin.
Soap and warm water and a tooth brush are used to clean and rinse the crystal. In crystal shops clean compressed air is used to dry the, crystal and electrodes, but for home grinding lintless cloth from an optician's, or at least a. clean towel, can be used.
The equipment necessary for grinding a crystal blank to frequency. A piece of plate glass and a "button" of the same material are essential. The "quick-change" adaptation for the crystal holder is a convenience. Not shown, but also convenient, are a small paint brush for spreading abrasive and a toothbrush for scrubbing.
Fig. 2 - The 1/2 by 1/2 inch electrodes used in modern crystal holders, showing the lands at the corners between which the crystal is firmly held.
Present-day electrodes have raised lands on each corner, as shown in Fig. 2, and the crystal should lie at least half-way across these lands and should not be larger than the electrode. The electrodes should be cleaned as carefully as the crystal. Before final assembly both crystal and electrodes should be handled carefully by the comers or edges after their last good scrubbing.
How to Grind
The actual grinding is done as follows: Spread the 303 abrasive over am area about a half inch square on your lapping plate, wet the brush, mix water into the spot and spread the abrasive over the lapping plate. Always keep the abrasive moist. Take the button and put a drop of water at its center, and press a dry blank over the drop of water. There should be just enough water in the drop so that it squeezes out under the edges of the blank, where it is wiped away. Place the button, blank down, on the emery and put the index finger in the center of the button. Use just enough pressure to move the button in a figure-8 pattern. This motion is used because it seems to balance the hand directly over the button and helps keep the blank flat.
After grinding through ten or fifteen "8s" the blank should be rechecked for frequency and activity. The blank's activity is a term used in crystal making to describe how strongly a crystal will oscillate. In ham gear this might be indicated by the magnitude of the dip in the plate current, grid current to the next stage, or rectified grid current in the crystal oscillator. It is nearly impossible to tell how much change in frequency will occur during the grinding of a crystal, because pressure on the button, the amount of abrasive, and the area of the "8" all will vary the frequency. The frequency change probably will be between 200 and 1000 cycles per "8," using a 7-Mc. crystal. The crystal can be moved along faster as the operator becomes more familiar with the technique, but for the beginner frequent checks of activity are in order so that any drop can be corrected.
To grind a crystal successfully the activity must be good when the crystal is brought to the desired frequency. There are several ways to raise the activity. Assuming that, with careful grinding on a flat plate with a flat button, the two faces of the crystal are parallel, the major cause of low activity will be dirt or moisture on the crystal or electrodes. Before checking activity the crystal should be scrubbed carefully with the tooth brush, using warm water and soap. Wipe the crystal clean and be sure that the electrodes are clean and dry. If the activity is still down the next thing is to bevel all eight edges of the crystal, as shown in Fig. 3. The beveling can be done with either fine or coarse abrasive, but is usually more effective with the coarse. Beveling, incidentally, will also raise the frequency because of the quartz ground off during the process.
Although beveling will usually improve the activity, another method - and probably the simplest - is to change electrodes. The land heights on the electrodes have a critical effect on activity. This is attributable to two things: the capacity across the electrodes, and the reflection of supersonic waves which are given off by the crystal and reflected by the central portion of the electrode. These supersonic waves are caused by the flexure or rise and fall of the center of the crystal as it vibrates in the shear mode. If the center of the crystal becomes too high and the lands are so low that the center of the crystal touches the center of the electrodes, the crystal will stop oscillating.
|Another view of the crystal holder, showing the assembly.|
The last step and the most drastic method of raising activity is to edge-grind adjacent edges, as in Fig. 3. This grinding is best done with coarse abrasive and should be followed by a slight bevel to remove any chips which may remain. The author uses the figure-8 motion for both beveling and edge grinding to keep the bevel even on all sides and to keep the sides parallel. By checking the crystal frequently, a drop in activity can be corrected by the above methods. If the crystal is ground too far and goes completely dead, the frequency may be too high when the crystal is again active.
Since the author is a ham, the gear in which the crystals were chocked was also taken from the 1945 Handbook. Two crystal oscillators were built as recommended, one a 6L6 pentode oscillator and the other a 6L6 Tri-tet. The circuits used are given on pages 97 and 98, the only difference being that in the author's gear the plate voltage was 450. These two oscillators were arranged so that either one could be used to drive an 807 final which was also built as per ye old Handbook. Thus, with no trick circuits or special gear the crystals were ground just as any ham probably would go about it. With a 150-ma. pilot bulb to protect the crystal the rig was put on the air. Allowing for a rusty fist the signal was cleancut and all that could be asked of a crystal. Seven-Mc. crystals were used. With the pentode oscillator on 7 Mc. and the 807 on 14 Mc. the rig worked fine -with a dummy load. In the Tri-tet oscillator the 7-Mc. frequency was quadrupled and the 807 worked as an amplifier on the 28-Mc. band. Many contacts were made and reports indicated that the signal was still keying well and according to the best ham technique. Standard 14-Mc. crystals were used in the pentode oscillator with the 807 doubling to 28 Mc. and still putting out a good signal. For the average amateur, 14-Mc. crystals are rather difficult to grind as they are touchy things to handle, but it is not difficult for any ham to grind his own 3.5- and 7-Mc. crystals. When a crystal was on frequency it was tried in the rig. In all cases each crystal was made to work by the recommended methods for improving activity and obtaining the necessary drive.
Finally, there should be a word of warning about out-of-band operation. The mounted crystal should be tested for stability of mechanical assembly before the transmitter is put on the air. This can be done by rapping the holder several times on the edge of the table. Another method is to drop the crystal from a height of about a foot on to a hardwood surface. If any change in frequency can be observed the holder should be taken apart and reassembled. In modern holders the spring may cause instability if it is not seated properly. If good frequency-measuring equipment is not available, do not try to crowd the edges of the band. Changes in frequency in the order of 500 cycles can occur between various oscillators, and smaller changes are possible in tuning. If you must crowd the band edge, get another ham who has accurate measuring gear to check your frequency. Grinding your own crystals can be lots of fun, and you can have the freedom of a v.f.o. without the danger of a pink ticket.