Introduction

Some CW operators that use keyers, often think that they might like to try "Touch Keyers". Touch Keyers are a bit different then than keyers with mechanical paddles. There is no movement of the paddle. All that is required is a light touch. There is often a sensitivity adjustment, but not always.

Touch Keyer - YC5NBX

The schematics below are for a Keyer along with a Touch Paddle interface. While the main interest here is the Touch Paddle interface, I have also included the Keyer. Both designs are attributed to YC5NBX, however, I could not find a name to associate with that call sign. Also, the schematic that I obtained said it is Iambic. While it does work fine as a keyer, I can't see where it can be considered Iambic. Squeezing both paddles at the same time would only send a string of dashes. If you need more information on what Iambic keying is, read this PDF on Iambic Keying - Debunking the Myth.

The Touch Paddle interface contains two separate interfaces. One for the Dot input and another for the Dash input. The paddle sensitivity is adjusted by the Drive Sensitivity control. This determines how sensitive the paddles are to your touch. This section could be used with other keyer to add touch capability.

The power supply for the entier unit is at the bottom of the schematic. It is intended to run off of 13.8 VDC, which is a common power for most transceivers. If there is no interest in the Keyer section, ignore the power wiring for U2, U3, and U5. Those ICs are part of the Keyer.

This page is the Keyer schematic. It may seem that there is a lot of extra circuitry, everything is there for a reason. This keyer is intended to work with a SSB transceiver. The section in the upper right, labeled PTT Output, is designed to provide the PTT signal with a delayed turnoff. Each time a dot or dash is entered, the delay is extended. The control VR3 is used to adjust that delay to your transceiver. At the same time as the PTT is being generated, the Tone Generator is also keyed through Q7 and relay, K1. The output of the relay can be sent to the microphone input, MIC, of the transceiver, or it can be sent to a SPEAKER. The Tone Generator is adjustable from 150Hz to 12KHz.

If this keyer was to be used with a CW transceiver, that is directly keyed, the PTT section, Q5, Q6, U4, and their associated components, would not be required. Also the Tone Generator, U5, and its associated components, would not be required, unless you need a side tone. In which case, the output of the relay, K1 would go to a SPEAKER and the MIC input would not be required.

There is two minor errors on the drawing. In the original drawing:

• LED D9 has the Anode connected to U4 and the Cathode connected to +12 Volts through R24. This is backwards and would prevent the LED from lighting.
• The Reference Designator D10 is used twice. The diode across the relay, K1, should be R11.

Touch Paddle Keyer - Rev. Michael F. Windolph, W0OGX

Initially I had some trouble finding the source of this keyer. The original article was in the March 1978 issue of 73 Magazine. That was a great help because the schematic that is floating around the internet was a bit fuzzy, making it difficult to determine some pin numbers. Plus, all of the transistors and diodes do not have a part type listed. Although, general purpose simiconductors would work. Well, it turns out that that is what the author suggested.

Since this page is really about Touch Keyers, I will start with the touch input for this keyer. On the right is the schematic. The outputs simply connect to the keyer where the Dot and Dash paddles would normally be connected. While it isn't mentioned in the original artical, this input setup could be used with almost any keyer.

The original article provides a sketch, and a brief description, of the Touch Paddle. The base for the paddle is a piece of stiff plastic about 3/4" wide. It needs to be long enough to reach into the the keyer enclosure and reach a bracket. Then, two pieces of thin aluminum, one for each side of the piece of plastic, are cut and glued to the plastic. The sketch in the original article shows some offset bolts, needed to make contact with the aluminum sections. It's not real clear to me how the finished paddle is secured and connected. So you will just have to figure it out yourself.

Of course, don't forget the +5 Volt connection.

This next drawing is a redraw of W0OGX's original article. It only included the oscillator (top left), the code generator (bottom), and output interface (top right). The red numbers, 0 and 1, are used to indicate the logic levels when the keyer is powered up, and neither paddle closed. These levels can give you a good idea of the keyer's operation.

The oscillator is normally stopped and only runs when either of the paddles are closed. When the Dot paddle is closed, U3B-8 gets set to a 1. This raises the level on the Cathode of D3, which enables the clock. That same signal goes to U2B-12, which propogates through to U2D-11 and U2C-10. The output (CW_OUT) goes to a 1 for one clock cycle. The CW_OUT signal is then used to clock U3B-8 back to it's idle state, 0. This disables the clock by lowering the Cathode of D3 to 0.

When the Dash paddle is closed (set to 0), diode D5 forces the Dot paddle input to go low also. This sets U3B-8 to a 1, enables the clock. With U3A-2 set to a 1, the flip-flop can be clocked. The output of U2D-11 starts to generate a Dot and enables U3A. At the end of the Dot, U3A is clocked and fills in the rest of the Dash. When the Dash is complete, U3B and U3A return to their idle state.

The drawing below show the Sidetone Generator and Power/Ground wiring. The Power/Ground wiring includes a 12 VAC transformer, bridge rectifier, filter capacitor, and 5 Volt regulator. If you have +5V DC available from another source, that could also be used. As shown, the keyer will need about 250 ma at 5 volts. Although not shown on the schematic, it is recommended that 0.1 µF capacitors be placed across the VCC and Ground for each of the ICs. This will help prevent noise spikes from causing erattic behaviour.

If you don't need the Sidetone Generator, it can be eliminated, saving on the current requirements. Often, the transmitter has a built in side tone generator. The original article also has a possible alternate Sidetone Generator, that uses only three transistors. This could save a lot of current.

Table 1. Debugging Chart for the Super Deluxe Keyer
With power applied, but key not contacting, check for the following:
	High (near 5 volts)	Low (near zero)
IC1	Pins 1, 2, 5, 6, 9, 10, and 11	Pins 3, 4, 8, 12, and 13
IC2	Pins 1 (grounded when Dash is sent), 2, 6, 9, 10, 11, and 13	Pins 3, 4, 5, 8, and 12
IC3	Pins 1, 5, 6 (grounded when Dot or Dash is sent), 9, and 13	Pins 2, 8, and 12
With power applied and either Dot or Dash paddle grounded:
IC1	Pins 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, and 13 vary at an intermediate voltage while keying, depending on keying speed.
IC2	Pin 1 goes low on Dash. Pins 4 and 12 go high and remain thus while keying. Pins 5, 6, 8, 10, 11, and 13 vary at an intermediate voltage while keying.
IC3	Pin 1 and 5 vary in voltage with keying speed. Pin 6 goes low when Dot is sent. Pin 8 goes high and remains there while keying. Pin 9 goes low and remains there while keying.

The table on the right was provided by the author, to help novice builders troubleshoot their keyers. The top part of the table provides the logic levels for the keyer in the idle state. The idle state is with neither paddle closed. The builder can check for these logic levels with nothing more than a VOM. Preferebly, the VOM should have a high Ohms/Volt rating.

This is a good idea, but there is a error in the original article's table. Under the settings for "With power applied, but key not contacting", only IC1 and IC2 are listed. It looks like IC3 logic levels are missing. But that's not it. Actually, IC2 logic levels are missing and the IC2 label should be IC3. The table on the right is the corrected version. This error is also detailed in the error section below.

There is some minor errors on the original article. Some are not really errors. They are just things I don't like about the original schematic. :

• Some of the schematic annotation is so fuzzy that it can't be deciphered. But the other drawings in the original article have helped clear things up.
• IC3 is shown as a empty box with no annotation and wires attached. In reality, it is a dual flip-flop. Each flip-flop should be shown separately and annotated properly.
• IC1-B has an incorrect pin number on the gate input. It is listed as pin 7, but should be pin 4. Pin 7 is ground. It is correct in the wiring layout.
• There is no indication of the type of transistor, diode, or FET. But in the original article, the author makes it clear that almost any type can be used.
• Some reference designators are missing.
• Reference designator TR6 is used twice.
• In the original article, the schematic shows the Anode of D2 and D3 are connected to the junction of R4, C2, and R6. That is correct, but the Keyer Wiring Layout (Fig. 2.) only shows the Anode D3 connected to the junction of R4 and R6 and the Anode of D2 connected to only C2. That is incorrect.
• Power/Ground tie-ups are not shown for any of the ICs. They are mentioned in the schematic drawing label, in the original article. I feel that all wiring should be on the schematic, not just a side note.
• In the Debugging Chart, under the settings for "With power applied, but key not contacting", only IC1 and IC2 are listed. While it appears that IC3 is missing, actually IC2 logic levels are missing and the IC2 label should be IC3. The table in this document is correct.

An Ash-Proof Keyer Paddle - Roy Lewallen, W7EL

The original article for the Ash-Proof Keyer Paddle was in the August 1981 issue of QST Magazine. Roy's remark about Ash-Proof really means that Touch sensitive paddles eliminate any of the contact problems that mechanical paddles might have.

I think of it as a Touch interface because, it can be use with any keyer that relys on the Dot/Dash inputs being pulled from a "high" logic level to a "low" logic level. I included a description for the paddle itself and adjustments, directly from the original article because I couldn't figure out how to say it any better then Roy did.

Construction - The mechanical portion of the paddle is constructed as a three-layer sandwich as shown in the drawing. Ordinary pc-board material is adequate and readily available, but other materials can be used if desired. The center layer is needed to shield the two sides from each other and is connected to ground.

Paddle sensitivity to touch is inversely related to the capacitance of the paddle assembly itself. Using ordinary 0.065-in. (1.7-mm) glass-epoxy board with each side of the paddle having an area of one square inch (645 sq. mm), sensitivity is more than adequate. Increasing the paddle area or decreasing the material thickness will reduce the available sensitivity.

U2 should be mounted close to the paddle to minimixe lead lengths. Each comparator input should be laid out symetrically to aid balance. Layout is other wise uncritical. My paddle circuit, along with a simple keyer, is built on perforated board and mounted in a small metal box. (Roy Lewallen)

Adjustment - Advance the SENSITIVITY control (R2) until one side of the paddle operated spontaneously. Adjust the appropriate trimmer capacitor until operation stops or changes from dot to dash or vice versa. If the keyer operation is reversed, adjust the other trimmmer; otherwise, increase the sensitivity. Repeat this procedure until R2 is at a maximum and no spontaneous operation occurs. R2 may then be adjusted to suit your personal tastes. No readjustment of the trimmers is necessary. You may want to experiment with the SENSITIVITY control to find the setting that feels best for you. (Roy Lewallen)

Touch Paddles - Peter-Heinz Guenther, DL2AWA

I looked around the internet, but couldn't find anything on this design, other than a clipping of the schematic.

Like some of the previous circuits, I think of this design as a Touch interface. It can be use with any keyer that relys on the Dot/Dash inputs being pulled from a "high" logic level to a "low" logic level.

While there isn't any explanation, I can see from the drawing that there are three metal plates involved. There is the plate that your hand should be resting on, and the Dot and Dash plates.

The jumpers (JP1 and JP2) are selected depending on the keyer you are using it with. If you are using this design with a Iambic keyer, you need to use the A:Squeeze settings (U1-A Pin 5 to U1-B Pin 6 and U1-C Pin 8 to U1-D Pin 9). This setting connects the inputs of U1-A and U2-C together.

But if you are using a non-Iambic keyer, you need to use the B:Normal settings (U1-B Pin 6 to U1-C Pin 10 and U1-C Pin 8 to U1-B Pin 4). This setting forms a latch so that the output lines that connect to the keyer can not go "low" at the same time.

Touch Sensitive Keyers - Geoffrey Walsh, GM4FH

Here are a couple of touch sensitive circuits from a March 1999 issue of RadCom. They are meant as simple replacements for a straight key. However, they could certainly be used with keyers. It is recommended that you read the entire artical as EG Walsh has some interesting insights into the ergonomics of touch paddles.

The first one, on the right, has minimum components (one operational amplifier, three resistors, and a diode) and consumes very little power (9 Volts @<4ma). The schematic shows a 9 Volt battery, but power (9-15 VDC) could be easily tapped from a solid state transmitter.

The touch contact is a plain brass button 5/8" in diameter, obtained from a haberdashery counter at a large departmental store, insulated from the box by a tap washer, and glued in place. It is easy to solder a wire to the ring at the back of the button. Earth for the hand is provided by the metal box that the button is mounted on.

The TX Output assumes that you are using a up to date solid state transmitter, where the key input is simply pulled to ground. The key circuitry should have no difficulty sinking the TX key input. D1 may not be necessary. I was put there to ensure that, too high a positive voltage, from U1, did not reach the keying circuitry of the Tx.

The inverting input of U1 (-) is held a little below the positive rail, by the resistance divider R2/R3. When the finger is not in contact with the touch contact, the Non-Inverting input of U1 (+) is held above this by R1. This causes the output of the Op-Amp to be positive and no keying takes place. When the finger makes contact with the touch contact, the voltage on the Non-Inverting input of U1 (+) falls below that on the Inverting input of U1 (-). This causes the output voltage to drop and activate the transmitter. Current will pass from the key socket into the operational amplifier, through Dl. The action depends on skin conductivity. Exceptionally dry skin may need to be slightly moistened, although this problem is rare.

The second version of the touch sensitive key has a Piezo Buzzer so that the key can be used for code practive. It also has a transistor output, which is capable of dealing with higher currents than the Op-Amp would have by itself. To drive the transistor, an inverting stage is needed on the output of U1A. A CA3240 Dual Op-Amp handles this easily and helps to keep the parts count at a minimum.

The power is off when S1 is in the central position, to one side power is on and buzzer is on, to the other side power is on and buzzer is off. When practising without being connected to the transmitter it is appropriate to use the buzzer. When transmitting, if the transmitter generates a side tone, it may be more pleasing to have the buzzer off. S2 enables the unit to be on continuously. This facility may be of use in checking power levels of the transmitter, VS WR, etc. The quiescent current drain is about 3mA and about 9mA when activated and the buzzer on.

AFC 1290k