I thought to put before you all my board construction technique. I realize my method is not for everyone and there are other methods

I use a copper clad board. The leads are short and the surface mount topology is reminiscent of ugly construction, but with the added feature of ease of construction due to treating the board as per pads. Anyone interested?

 

More detail? Any photos? To be honest I don't know what you're offering.

 

More detail is forthcoming....

 

I started building boards with perforated plastic as sold by Radio Shack. But these were sloppy. After a while I discovered copper clad boards, but I didn't know about etching them with chemicals. So I used a knife to cut away the copper I didn't want. Then I discovered Ferric Chloride and this made it easier to remove the copper, but I had to wait for hours -- sometimes overnight -- for the process to have occurred.

Several months ago I got a Dremel 4000 and it solved the time problem. I can now put together a board pretty fast. No more waiting overnight for the copper to be removed.

I anticipated this post a week ago when I put an AM Modulator together and so I took some pictures to document the process. The circuit is built around an LM1596 in a ten-pin can (TO99?). The only changes I made to the manufacturer's design was to a resistor and a capacitor such that it's lock point is centered on 455kHz.

 

The first thing to do was to label the nodes then draw a picture reflecting them in a practical physical arrangement:

 

Then, using the Dremel, I cut away the unwanted copper:

 

The next step is to protect the board from future oxidization. I have some boards left over from the nineties which I left unprotected and they are now dulled with oxidation. I believe this is especially undesirable with radio due to the skin effect, and this is why, I suppose, manufacturers paint their boards -- usually with a green enamel.

To do this I smear flux all over the board, liberally. Then floating the solder onto the copper is easy and takes only a few minutes for a board this size. Here's the tinned board:

 

Here's the completed product, except two BNC connectors I'm expecting in the mail tomorrow.

 

Not a bad idea. I suggest that you use double sided copper clad board. Leave the bottom side as a solid ground plane.

Secondly, start using SMD components. They will help to reduce noise and improve RF performance.

I do the same but instead of etching the boards, I cut away the unwanted copper with an X-ACTO knife. I cut two parallel lines and then roll the copper strips off, like rolling sod off the ground.

Here is an example of one of my projects:

 

Thanks Mr Chips I like your ideas about using double sided boards and SMD components. I shall do both as of my next order from the store. As for using an Xacto knife, I'll pass on that. I like my Dremel!

On edit after taking another look at your board I can say I really like those SMD components. One criticism, for whatever it's worth: Back in school I was warned against sharp corners, such as those square intersections you made; it had something to do with creating an antenna, I think.

When I was working at Systron Donner in Sylmar Ca., we prototyped a frequency synthesizer for a Japanese firm named ANDO. The design was good, but the software that designed the boards was not sophisticated enough to do the job. It introduced many unwanted mixers and antennas by accident which in turn produced so many spurious signals the synthesizer was rejected.

 

there is also a coper clad perforated fiberglas board available, with a 1/10 inch grid of holes. makes it a lot easier than having to drill holes.

 

Thanks PRS and MrChips for sharing your construction methods.

...
Then I discovered Ferric Chloride and this made it easier to remove the copper, but I had to wait for hours -- sometimes overnight -- for the process to have occurred.
...

I use ferric chloride, and after warming it up and agitating it only takes 5-6 minutes for a full etch. I use it full strength out of the bottle, it's very dark liquid almost black.

 

Thanks PRS and MrChips for sharing your construction methods.




I use ferric chloride, and after warming it up and agitating it only takes 5-6 minutes for a full etch. I use it full strength out of the bottle, it's very dark liquid almost black.

Well, that's the same stuff I used. I stored it under my kitchen sink this winter and it got really weak. I suppose this was due to the cold. I know that heat by way of heat lamps speeds up the process but now, with my Dremel, I don't have anything to worry about. But I respect the chemical way of doing this thing. So how long does it take you to etch a board with ferric chloride?

 

This is true, but when you have a node with lots of connections, I should think such a thing would be inadequate for rf.

 

Since we are on the topic of prototyping, here is how I do it.

I begin with some double sided PCB stock. I tin one side and then use a sheer to cut it into small squares (pads) and strips (busses). This gives me a stock of pads and busses that I can solder onto a piece of single sided PCB. If I need a pad I just tin the spot where I want it, hold the pad in place and re-flow the solder under it. See photos.

I have used this method for RF designs but cleaned out the old prototype box a while ago.

Mark

 

Nice idea with the Dremel... especially since I already have one

 

+1 for smd. You can get like 5000 smt resistors for a fiver from some online retailers.

 

Very innovative, I like that. Another method I just read of is using copper foil and cutting it into strips then gluing it to a fiberglass board with the bottom sheeted with copper. Grounds are made by drilling and feeding a wire through to the ground plane.

Your method looks like its good through 250 MHz or so. Have you tried it with higher frequencies?

 

I'm sold on surface mount parts. But tell me, what in the world does your signature mean? Reminds me of back when I smoked weed. Say that to somebody smoking weed and you'll give him a rush.

 

... So how long does it take you to etch a board with ferric chloride?

Only 5 or 6 minutes from putting the PCB in the tray to pulling it out etched. I use a flat tray about 4" x 6" and just enough FC to cover the PCB, about 1/4" of depth I suppose. Then I "slosh" the tray so the liquid moves over the top of the PCB vigorously.

The etching happens so quickly you can see it happen. I'm always amazed when people say it takes "hours".

It is important to make sure it is nice and warm, I put it in full sun on the hot concrete which would make it about 50'C but I also have a memory of putting it in the microwave (without PCB!) to heat it, before putting the PCB in.