Hi,

As suggested by SgtWookie and others I'm using dead-bug construction to prototype RF circuits, and I'm getting much better results than I did using stripboard.

I have a question, if I glue a strip of copper-clad board on top of the ground plane to use as a positive rail will this have a negative effect on the circuit in the same way that long leads add stray capacitance?

Sort of a cross between dead-bug and manhattan style - I find it easier to get a neater layout like this.

Example:

The circuit works fine, but I want to prototype in the best manner possible.

 

Two flat conductors with opposite polarity voltage applied facing each other is pretty much the definition of a capacitor. I'm sure you will not improve the performance of the circuit that way. Better to build up the sides and top to form a box that is connected to the ground plane. This will isolate RF to the interior. A port window or direct conductor pass through will get a signal into and out of the 'box'. If you look at any old hands HAM radio equipment you will find such isolation being used around RF oscillators and IF sections as well.

No need for a power plane of copper. Run the voltage with wires and use extensions of the ground plane to isolate any component that cause radiated interference in the circuit.

 

Two flat conductors with opposite polarity voltage applied facing each other is pretty much the definition of a capacitor.

That makes sense

No need for a power plane of copper. Run the voltage with wires and use extensions of the ground plane to isolate any component that cause radiated interference in the circuit.

Thanks for the advice

I find dead-bug quick as well to construct compared to stripboard

 

Hello,

Also keep in mind that sometimes the housing of the transistor can be connected to one of the legs (mostly emitter).

Bertus

 

Also keep in mind that sometimes the housing of the transistor can be connected to one of the legs (mostly emitter).

Thanks - you mean like the 2N2222 in the picture with the metal can?

 

Google up amateur Ham radio and related topics, look for the rigs they build themselves. Many of their build logs will show you very creative and innovative construction techniques for use in RF freq circuits. RF circuit design is a specialty and has mechanical constraints and requirements other lower freq and powerDC circuits don't have. I think you will be surprised at some of the items they employ in building a transmitter or receiver

 

Hi Kermit2,

Thanks - I've been doing that and find it fascinating.

 

Instead of sticking bits of board to the existing board, there is another method I have learned from a Japanese website. A tool is used to cut off an island in the form of a pad on the ground plane suitable for soldering or standoff.

I made the tool myself as one might be be able to buy it ready made.

I just use the tool as needed when I go along with the circuit construction.

Example(5Mhz to 10Mhz frequency doubler):

 

Hello,

You can also take a look at this thread about RF prototyping:
http://forum.allaboutcircuits.com/showthread.php?t=32886

Bertus

 

Instead of sticking bits of board to the existing board, there is another method I have learned from a Japanese website. A tool is used to cut off an island in the form of a pad on the ground plane suitable for soldering or standoff.

I made the tool myself as one might be be able to buy it ready made.

I just use the tool as needed when I go along with the circuit construction.

Great! I've seen people also use a Dremel tool to cut the copper cladding to make an island. Thanks.

 

Hello,

You can also take a look at this thread about RF prototyping:
http://forum.allaboutcircuits.com/showthread.php?t=32886

Bertus

Thanks Bertus, will do

 

When using metal can transistors, can I stick the metal can directly on the ground plane or do I need to isolate it?

 

That will depend on the transistor. Older TO-5 (now TO-39) transistors usually had the case connected to the collector. TO-3 and -66 transistors all had the case and collector tied together (case was the collector lead). Can't say about TO-18's.

A check with a meter can tell you if any lead is electrically bonded to the case.

 

That will depend on the transistor. Older TO-5 (now TO-39) transistors usually had the case connected to the collector. TO-3 and -66 transistors all had the case and collector tied together (case was the collector lead). Can't say about TO-18's.

A check with a meter can tell you if any lead is electrically bonded to the case.

Thanks.

I guess isolating the metal can from the ground plane can't hurt.

 

Amen. For the record, the 2N2222 always has the collector tied to the case, not the emitter. I know that little part very well.

 

Thank you Bill.

 

Dude, don't get too wrapped around the axle on simple air circuits. Use a plastic TO-92 case for your transistor. The capacitance of the strip of clad glued on the ground plane would be so high, it likely won't affect your circuit at all. If you are that worried about it, take a knife and gouge a groove through the copper clad, making sure it's electrically isolated from the ground plane and use that as your power rail. Seriously. You are building an air circuit. Every one of those long component leads are acting as antennas anyway.

 

Every one of those long component leads are acting as antennas anyway.

This means that all threads on a PCB and legs on components should be as short and small as possible. Correct? Wouldn't it be better to build it on a perf board, and cut the legs when soldered?

Which in turn means that RF circuits should be built with SMD. Correct?

But then it wouldn't be dead bug style...

Oh man... I need a great magnifyingglass/lamp....

 

Dude, don't get too wrapped around the axle on simple air circuits. Use a plastic TO-92 case for your transistor. The capacitance of the strip of clad glued on the ground plane would be so high, it likely won't affect your circuit at all. If you are that worried about it, take a knife and gouge a groove through the copper clad, making sure it's electrically isolated from the ground plane and use that as your power rail. Seriously. You are building an air circuit. Every one of those long component leads are acting as antennas anyway.

Thanks... I have tried both isolating "islands" of the ground plane with a cutter as well as sticking board pieces with superglue and both work well.

 

yes, with RF circuits, leads should be as short as possible. Technically, RF circuits should be in metal enclosures. SMD aren't necessary if you don't want to re-engineer the circuit layout. I've built many prototype RF circuits on perf-board and they work well. Etching a PCB is the most preferred method if you have the time and resources.