Hello, this is my first post here so I guess I'll introduce myself. My name is Matt and I'm an electrical engineering student, currently a sophomore. I recently got an Amateur radio operator's license which is where this project got its start. I'm an EE major but it will surprise you how little I know about circuits thus far. Anyway...

I'm putting together a morse code keyer, for those who don't know what that is, it is a device that a morse code key is connected to that creates a tone so you can hear what you are sending. It also puts the transceiver into transmit mode. It has a few more functions but that's basically what it does.

I am pretty much a noob when it comes to circuits, I mean I know how to solder and I've put together a few altoids can guitar amps and even wired a strat. My project will be a combination of 2 kits available on the web as well as an LM386 amp. The first kit is the actual keyer, the second kit is a capacitive touch circuit that will be used as the morse key to input into the keyer.

Kit 1: http://www.hamgadgets.com/store/product_info.php?products_id=89
Kit 2: http://www.cwtouchkeyer.com/P3W.htm
LM386 circuit: http://web.mit.edu/6.s28/www/schematics/lm386.htm

The Picokeyer uses 2.5 -5.5V and draws a max 2mA, and the touch key uses 6 - 14V and not sure about current draw. I'm planning to use a 9V 1A wall wart to power the touch key and the LM386 and then using a voltage regulator like the 78L05 but with a 3V output to power the picokeyer. The speaker connected to the LM386 will be an 8Ω ripped from an old pc speaker set. I have no idea if a 9V 1A supply is enough or ideal for this project so I would appreciate any opinions.

I'm using the lm386 because I'm going to put this into a plastic case and the piezo transducer the picokeyer uses will be too quiet to hear. The tone output from the picokeyer is a square wave. The kit maker says that if you integrate the kit with homebrew equipment that you should use a low pass filter to make it cleaner. (I believe Morse code tones can range from about 250-900Hz) Now I was wondering if the filter should go on the input of the LM386 or the output to the speaker. Also, would some sort of square-to-sine wave filter be better than an lp filter? I don't know what would go into converting a square wave to a sine wave but an lp filter would probably only have a small cap and coil right?

One more thing, sorry this is so long! I want to use a 100k pot to control the volume but should the pot go before or after the amp?

I will post updates and pictures throughout the project if anyone is interested, maybe even a video. Here's a quick video I found of the picokeyer: http://www.youtube.com/watch?v=660WwT63I_Q

Thanks in advance,
Matt KC9QNC

 

Congratulations on receiving your amateur radio license! I know from experience that it's a superb hobby, and code is one of the best parts.

You know, you could construct a Morse Code Keyer without having to buy kits, and save a lot of money. Here's a link that includes several schematics to do just that:

http://www.discovercircuits.com/M/morse-code.htm

As for the capacitive touch circuit, I remember reading a QST article of a guy who designed a Morse Code Keyer with capacitive initiation (it had everything you described). I'm not sure whether you're an ARRL member, but the schematic is available from that website if you search within the QST keyword selection. I suppose you could add the capacitive switch kit you found, though.

If you implement the filter to the input of your LM386 amplifier, you will get some attenuation of your original signal. Therefore, it might be best to add the filter on the output of the LM386.

Adding a pot for volume control will be within the LM386 circuit, as part of the gain control. Take a look at the LM386 datasheet and see what I mean; they have several circuits in which you can vary the gain of your amp (thus varying the volume).

 

Thanks for the quick reply Electronerd. I'm learning cw and I can't wait until I can make my first QSO, I find it exciting.

Yes I'm an arrl member, I'll search the QST archives for that. I know I could save myself a bunch of money by not using kits but as of right now I don't think I have the self confidence. I guess it's like using training wheels :/ I've never put together my own circuit before so I'll be using a breadboard first and then putting this onto perfboard I think, also using the pcb's included in the kits that I already ordered I know this won't be my last project so maybe next time I'll do it from scratch. This project was inspired by a Heathkit HD-1410 keyer that my school's club has. It's a cute little thing lol
here's the HD-1410:

EDIT:
I didn't want to double post. I found this circuit in the lm386 datasheet, if I replace the fixed resistor between pins 1 and 8 with a say... 100k audio pot will I be able to have the full range of gain from 20 to 200? Also what does the 10k pot on the Vin do? is that for adjusting the input level?

 

You wont want to use an audio pot, they change resistance over a non-linear scale. They are for changing the resistance to follow human hearing.

But, you do have the right idea. A LOG pot will give you a full range on the output level.

And correct again about the 10k input pot.

Its just like the gain knob on a mixing deck. You find a good clean signal that just clips for input, then use the 100k pot to adjust the output.

 

The first QSO is very exciting for everyone; especially if you have limited output power and you reach a DX station.

I wouldn't advise that you construct the capacitance circuit on a bread-board; the breadboard will induce more capacitance and change your sensitivity. If you can, solder it to a PCB like in the kit you provided earlier.

 

This is all really helpful! Thanks!

Now about the power supply, I'm thinking 9V 1A will be fine, the lm386 data sheet says it uses about 400mA on 9V, the keyer uses 2mA, not too sure about the capacitive circuit. I think these will all be wired in parallel inside the enclosure? Also I just realized the speaker I have is 4Ω, how would that effect the performance of the lm386 circuit compared to an 8Ω?

I searched the QST archive, couldn't find the capacitive touch circuit but I did find a bit on it. Someone wrote in about the use of the capacitive circuit near high RF exposure, they needed to put a choke on the dot, and dash outputs or else it would cause spontaneous clicks. I also found a complete schematic for a keyer, less the memory/contesting function of the picokeyer, but I liked how it looked and I might build that one in the future.

 

You will be happier with 6 AA batteries (or C or D) rather than a 9v "transistor" battery.

9v batteries have VERY little capacity.

 

Sorry for reviving an old thread but I'm building capacitive CW keying paddles and have something to add here in case others stumble over this or are interested.

Basically the "P3W CW Touch Keyer kit" referenced above is a complete rip-off at $29, and if you want to combine it with other stuff in your own design, even more so. It irks me that the seller of these kits has deliberately painted over the markings on the sensor IC's to make it harder to figure out how it's made. But not that hard if you know what you're looking for!

Atmel has a range of capacitive touch sensor IC's with identical or similar properties to the ones used in that kit. The cheapest ones that are suitable are less than $1 each, but those are SMD devices. If you don't mind using SMD, the AT42QT1010 at about $.75 is a good choice. datasheet Also it just requires a single external capacitor in addition to the chip itself. So a replica of the kit above would cost in the region of $5, allowing for a low-current voltage regulator. This can drive the input to your keyer stage directly, (or the GPIO pins of a microcontroller as in my case), with no need for amplifying transistors.

Thomas