These notes have been prepared by G3IRM to give you some idea how coherent c.w. operates. As you all will know, morse code consists of dots and dashes arranged to represent letters, figures and punctuation marks. The basic element is the dot which has a predetermined length depending on the speed of transmission. High speeds require short dots. Hand sent morse code and for that matter code sent with semi-automatic and automatic keys is random in nature and, even if the characters are correctly sent, the spaces can vary considerably.
The standard speed for coherent c.w. adopted by Ray Petit W7GHM when he designed the system was 12 words per minute. In actual fact it was the dot length which he decided should be fixed at 100 milliseconds. An element space was, therefore, also 100 milliseconds, a dash and character space were each equal to 300 milliseconds and all other spaces were multiples of the same 100 milliseconds. This timing is still in use today and is available on both the COHERENT and PCW programs available for coherent c.w. By connecting two computers together via the RS232 interfaces, one running COHERENT and the other running PCW, perfect communication can be achieved proving their compatibility. Only transceivers need be added at each end for on-the-air communication.
Coherent c.w. depends entirely on accurate timing based on the element length. If you imagine two square waves both identical in timing but derived from separate oscillators and having the mark length equal to the space length and both these equal to 100 milliseconds it will be obvious that, if they both start together, they will stay in step indefinitely if the clocks generating them are perfectly stable. The original hardware coherent c.w. system was based on this principle and highly stable standard oscillators and transceivers were required at both ends of the contact. The difficulty of providing these is the major reason why coherent c.w. did not become popular.
Modern computer programs have changed all that and now only reasonably stable transceivers and no frequency standards are required but it is obvious that the more stable transceivers and computers used in the system are the better will be the results.
I must now return to our two square waves. These can be looked upon as strings of dots each dot and space being of a definite length. If we now arrange for one station to send a string of dots these will only be received accurately if the timimg at the receiving end corresponds to our identical second square wave. Assuming that perfect timing exists at both ends the only thing that will stop correct reception is the relative phase of the two square waves. If the rising edge of one square wave occurs at exactly the same time as the rising edge of the other the signals will be sychronized (coherent) and all following transitions will correspond. Using the original Petit hardware filter this was achieved by the transmitting station sending a string of dots and the receiving station adjusting a phasing control until the string of dots was received clearly and distinctly. Once this was done and, assuming perfect timing, the stations would stay in step indefinitely. Dots and dashes transmitted would be received at the other end of the contact and as spaces no matter how long were all multiples of the element length, a dot or dash received after a space would still occur on the rising edge of the receiver clock. This is a somewhat simplified explanation of how coherent c.w. works but it will give you some idea of the principle behind the system. Let me repeat that in order to ensure correct operation all oscillators had to be exceptionally accurate and, above all, stable.
There are two coherent c.w. programs available. The first is called COHERENT and was written by Bill de Carle VE2IQ and it can be obtained from him. The other is PCW which was adapted from Bill's program and rewritten by Ernst Schroder DJ7HS to appeal to the operator rather than the experimenter. This is the program you will find on this disk under the name PCW.
The need for stable oscillators has been overcome in both programs. Automatic synchronization has been included so there is no need to have phasing controls. If the program receives a string of coherent c.w. dots it will automatically synchronize. Computer clocks and all but the best transceivers can drift a little and provision is made in both programs for the automatic correction of frequency drift. In COHERENT this is done by feeding pulses from the computer to the up/down frequency control lines of suitable transceivers. The recent Kenwood models such as the TS850 and TS450 are ideal as these can be controlled in steps of one hertz though models made by other makers may also be suitable. As everyone does not own such a transceiver DJ7HS decided to make the software carry out automatic tuning. No up/down lines are therefore required when using the PCW program.
DJ7HS has provided sufficient notes and on-screen help to enable you to run the program. It can be used to send c.w. as well as coherent c.w. (but only at the three speeds provided) as they are in fact exactly the same so far as transmitting is concerned. This is done by simply connecting the computer to the key input of the transceiver. Bear in mind though that the output of the computer is at RS232 levels and an interface may be needed to connect this to the keying line of the transceiver. A single insulated gate field effect transistor is all that is normally needed. Consult your transceiver manual. The program can only be used to receive coherent c.w. if the VE2IQ interface board is included between the receiver output and the computer.
Any transceiver will receive coherent c.w. as it is exactly the same as c.w. with the exception that it is timed extremely accurately and is, therefore, "perfect" morse code. This means that anyone hearing it can easily assume that ordinary c.w. is being sent. If you are prepared to receive the signal from the other station as c.w. you can, of course, make contact. In order to indicate that you can operate coherently it is necessary to include something in your call to this effect. You can send a string of dots (by pressing $) followed by CQ CCW DE callsign and possibly more dots. This will enable another coherent station to synchronize and call you. If you do not send the string of dots the calling station will have to send them to enable you to synchronize.
You are sure to receive calls from c.w. stations even though you indicate that you are using coherent c.w. They either will not know what CCW means and assume that you have a funny way of calling CQ or they may be curious to know what CCW is all about. Please take the opportunity to tell them.
Unlike some of the digital modes where it may be a little difficult to decide which type of signal is being received, coherent c.w. is easy to recognise by its perfect timing and string of preliminary dots. Even so you will have difficulty finding stations as there are not many stations active at the time of writing. For some time now it has been the practice of coherent operators to use one frequency. Various frequencies have been tried. At one time frequencies one kilohertz up from the usual QRP frequencies were used as it was thought that QRP operators had the most to gain from coherent c.w. and they may be tempted to make enquiries. This did not prove to be the case. It is now suggested that twenty kilohertz up from the bottom of each band be used. There is sure to be some interference whichever frequencies are chosen but coherent c.w. with its accurate timing and very narrow bandwidths stands a good chance of being heard. For the time being please use these frequencies and watch for any announcement of changes in the Digital Journal. I will try to ensure that other magazines are notified of any changes.
I hope you will enjoy using coherent c.w. and that we will have more activity in the near future.
Details of activity or any other items of interest are always welcome for inclusion in the coherent column of the Digital Journal. Please send them to me:
Peter Lumb G3IRM
2 Briarwood Avenue
Bury Saint Edmunds
Suffolk IP33 3QF
United Kingdom
| back to DJ7HS' download page | This text is copyright © Peter Lumb, G3IRM, February 1995 |