Homebrew 3 Channel Pure SIN Generator ( 0 ... 60MHz )

[Yucca]

Yucca , I hope you wont be stupid enough to offer this for free , after all that hard work .

You know theres a potential market for this ...

If somebody can only copy it , then he must buy it .

Dankie, I've pm'd you with more info but I plan to charge a modest amount for a preprogrammed chip with full colour laser printed build instructions and sources for all the components at the cheapest price, paypal through ebay. Existing customers will then be offered further firmware enhancements at a much reduced rate. It will make it easier for someone to build, just plug and play really. Also it will give me incentive to greatly enhance the software and produce a better product that will make a good peice of equipment with very high specs for anyones bench.

[stprue]

I'll buy one 

[Yucca]

@Yucca,
Good job. This is quite the addition for any future attempts. You will be pleased. I am digging deep into my controller failure. Seems the BEMF or shockwave burned all my outputs of the uproc.
I am writing a program to toggle the output lines in a standalone mode. No usb comm. I am locking this little bugger in a cage and gonna poke it with a sharp stick!
No caveats. But I must relay this bit of advice. I put all my current efforts and resources into this device. Don't have a backup plan as in other types of test equipment to do this type TPU testing. I have shopped and I don't like the pricing behind the LF & HF multichannel signal genny with synchronization capability.
So this controller route is good. Jason did implement alot of safety features. But nothing to guard against user misconfigs, like putting a feedback BEMF at the SSR gate drive that feedsback into the controller with no firewall. And I have a firewall safety that I built 2 years ago sitting there unattached! Argh. In my zeal gained from other builds and tests I ran roughshod over this new build and cataclysmically denied myself the favor of being peaceful in my attempts. So let this be a lesson to ya! I do all the blowin' up stuff here. LOL.
I am in code land and that is very dark and addictive real estate.

--gk.

GK, I hear you! codeland is my favourite sanctuary.

Ever since I fried a 50MHz analog scope when trying to look at a cars ignition timing when I was a teenager, ever since then I am more careful.

I have thought about different schemes to guard against fast back transients should they occur. In my initial pure sin experiments I don't expect major surgeback through the oscillators, but when I go to square, well things might get rough.

Having said that, even with initial sin build I will definitely be putting isolation into the feedback path:

[1] signal in

[2] rectify

[3] smooth

[4] bleed via 1ohm 5w resistor to ground (to give a DC voltage proportional to output current)

[5] tie high side of bleed resistor with neon and avalanche diode in parallel to ground (safety path to GND on overvolt)

[6] optoisolator

[7] log opamp to give compression and so more range

[8] then into the microcontroller ADC to calculate realtime RMS power and display as necessary

I suppose it wouldnt hurt to put neons and parallel avalanche diodes to ground on each oscillator channel also, negligable capacitance is added, whether they would sink fast enough to protect my RF output amplifier chips, I don't know? Maybe you could try this after you fix your uProcessor.

[giantkiller]

Famous last words: I'll be more cautious next time?
I have pulled all chips and am using the board as a holder for just the uproc. I have tried 3 of the 4 uprocs. All the same results. With nothing but caps, diodes, + a DC2DC cnvtr the processor runs and communicates. The voltage is +5v and fine. With one uproc left that hasn't been plugged in I need to program it in another setup in case this build is breaking uprocs. Eek! I buy more, but still.

--gk.

[Mannix]

I dont know if this is useful, but the cro i fried with my 3 channel ring had so many faults that seemed un related to the input channel. the faults were in the vertical amplifier,power supply and timebase ..separate,unrelated  faults it seemed ..I had a manual . I never succeeded in fixing it. It was a nice anritsu 100 mhz

This cat we chase sees pn junctions as cotton candy..so it seems