Thank you SLK001 - I'll try it
Thanks for taking the time to help - I did what you suggested and there's a slight improvement. Good enough to use this setup to hook it to the ultimate stage anyway - pretty sure I can get 5W out of that. Also, this 'hideous' board is hanging in the air with no shielding, being tested - it will go into a metal enclosure next and I hope that will tame it still further.It is both. Just estimate the inductance between the ground points. For instance, a via has approx 1.5~2 nH by itself. That's why I suggested more vias. They would be in parallel, so the effective inductance would be reduced by 1/N, where N is the number of vias.
Here is a suggestion to try to salvage your current board. The red lines are 16 gauge solid copper wires soldered from ground to ground. The red rectangles are additional 1000pF bypass caps.
View attachment 103113
I have 0.005" copper shim stock that I cut into 0.4" strips, then bend into a U shaped channel, slip this around the edges and solder both sides.Here's a little trick...a bit hard to do on a board like yours that has a "margin" around the copper. However, on double sided PCBs, I'll often just run a bead of solder around the whole outer edge, shorting out the upper and lower cladding. It's a lot easier than creating a whole bunch of vias. Of course, you don't want to do this for production boards, but for one-off projects, it works great.
Why NOT?I also would NOT put thermal reliefs anywhere on the board.
Each one adds a small amount of inductance to the parts that sit on a relieved pad. The reliefs also impede current flow, thus increasing your ground impedance. On low frequency stuff, thermal reliefs generally do not cause much of a problem, but on RF circuits they DO affect the results and can cause severe problems. Some parts of the circuit can be adjusted to compensate for the thermals, other parts cannot.Why NOT?
Certainly you're right - however I have just got the entire thing up and running at a total power of about six watts and it is giving me a nice stable plasma inside the tube - I would like to know where the harmonic is coming from (think it could be the scope even?) but I have arrived at where I wanted to be and am a happy camper!When I look at your signal output, I see harmonics. And that's the reason your waveform isn't uniform peak to peak. If you zoom your horizontal time base out, the true wave-from should begin to emerge.
I still contend that your NE85633 is the source. Using that device is like hitching a thoroughbred horse to a plow. It will probably get the job done, but you will be exhausted reigning the horse in all the time. Your output will reflect your problems.I would like to know where the harmonic is coming from...!
Keep in mind a plasma itself is a very non-linear device! Even if you feed it with a pure sine wave, plasmas will generate harmonics on their own. Depending on the degree of coupling to the plasma, this can be easily reflected back into the amplifier's tank circuit.Certainly you're right - however I have just got the entire thing up and running at a total power of about six watts and it is giving me a nice stable plasma inside the tube - I would like to know where the harmonic is coming from (think it could be the scope even?) but I have arrived at where I wanted to be and am a happy camper!
If anybody is interested I can post a picture of the final setup (all in a single box and ready to be redone on a single, smaller board, with SLK001's excellent grounding advice implemented) showing the lamp shining brightly and putting out lots of 123.6 nM photons! Strong ozone smell.
Gentlemen, your help has been very valuable - thanks!
I still contend that your NE85633 is the source. Using that device is like hitching a thoroughbred horse to a plow. It will probably get the job done, but you will be exhausted reigning the horse in all the time. Your output will reflect your problems.
Up to around the 1990's, the maximum oscillating frequency for a bipolar transistor was 1-1.5GHz. Then HBT (heterojunction bipolar transistors) came along and pushed the gain-bandwidth product (ft - that's f sub t) up by almost two orders of magnitude (up to 70GHz). Your device has an ft of 7GHz, so it is really chomping at the bit when held to only 100MHz. Maybe a device like the older 2N918 would give you better harmonic response. Just remember that your oscope is acting like a low pass filter when viewing the waveform.
Keep in mind a plasma itself is a very non-linear device! Even if you feed it with a pure sine wave, plasmas will generate harmonics on their own. Depending on the degree of coupling to the plasma, this can be easily reflected back into the amplifier's tank circuit.
Hi Spotty:I have also had a bit of prior experience - we marketed an instrument having a 5W inductively coupled plasma tube (low pressure) back in the 1980s. Although the plasma obviously has a strong resistive component I am certain you're right that it has a fair bit of reactance too. I'd like to see an equivalent circuit but an empirical approach has worked so far without the actual need for knowing that and I have quite enough fish to fry as it is. Thanks for your interest.
by Duane Benson
by Duane Benson
by Jake Hertz