concepts-easy.......application near impossible.

This area is for discussions involving any fusion related radiation metrology issues. Neutrons are the key signature of fusion, but other radiations are of interest to the amateur fusioneer as well.
Post Reply
User avatar
Richard Hull
Posts: 12851
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

concepts-easy.......application near impossible.

Post by Richard Hull » Mon Jul 09, 2001 4:06 pm

This is a post related to nuclear instrumentation concepts aplication and, ultimately, repair.

Over the July 4th holidays, I did a lot of Gamma spectometry with my canberra 8001. Ultimately, and to my horror, a large chunk of the spectrum disappeared from the 0 kev point out to 117 kev. I had to stop everything and take the lid off the beast and pour over the prints schematics and block diagrams to effect repair.

Having been an electronics engineer by profession for the past 35 years and an electronic tinkerer for almost 48 years, I looked at this as just another repair.

Opening the leviathon showed it to be 100% TTL/linear opamp based (1970's technology. There were several monster circuit boards and thousands of components. This was a frightening thing after all these years of de-training in board swapping and single 100 pin chip replacements. I grabbed hold of myself and realized I was back at a point when I was a junior engineer in the pits with the technicians back in the 70s.

The key is, first, the conceptualization of what is going on in the instrument. This is almost always simple. The MCA or multichannel analyizer basically looks at the roaring pulses from a PMT/scintillator's output at about a 200mhz rate measuring the analog pulse height of each, assigning a digitally encoded channel number to the value of the energy of the pulse and incrementing the data in this channel number's memeory by one for each pulse counted. It is then syncronously marched out onto a CRT back in an analog format of relative pulse count on the Y axis and channel number or KEV energy on the X axis. Sounds simple, but in the 1970's it was a task for hundreds upon hundreds of TTL logic chips and a plethora of op-amps.

Here is where the application is brought to bear. The unit consisits of a CPU type board that is composed of individual logical elements designed to control the signal flow of the system, a memory board with associated logic and timing circuits, a CRT board with sync circuits and D to A converters, two power supplies (one low voltage and one high voltage), an SCA or single channel amplifier and the key component the ADC or analog to digital converter board. The circuit diagrams, alone, are a 1" thick stack of 11 X 14 pages!

Where to start??!!

Fortunately, the Canberra folks knew the jungle they created would be daunting for a tech to just pick a bad component out of the 10,000 passive and active components within the 100 lb monster. So, they created and about 22 pages of block diagrams. These break the timing and control logic into mentally managable chunks. From here a good tech will get a "feel for what is going on (or supposed to go on). Unfortunately, they supplied no voltages or waveforms. (bummer) You have to be a electronics type to sort of know what should be there.

Repair is a fearfull thing when you are sitting there, late at night, even with all the prints, schematics and block diagrams. You just know that it is one component which has laid th' beast low.

I spent about 4 hours getting my sea legs on the concept of the original design enegineers pouring over the block diagrams. (application) From this base or core level understanding, the repair attack was strategically planned.

I checked the 15 different power voltages. (always a good bet) They were all OK. (couldn't be that simple)

After about another 7 hours of checks, I determined that all the base band levels, memory ICs, and the CRT circuits were good. Into my third night, my 10th soda and 5th bag of fritos I found that the problem was isolated to the ADC board which, by nature, would be a bear to troubleshoot.

I was really down to a mere 500 compnents! it was now a doable thing! Once homed in like this, you gotta' keep your wits or you will waste time. Since the first part of the band was missing (symptom) I checked the base band ref. voltages, all were good. I next checked the upper and lower discriminators and they worked with injected signals from my NIM ortec pulser. The problem now appeared to be related to the lower level set point reference and this voltage was good!! However, the on-board set potentiometer was not passing the voltage! A DAMNED CIRCUIT BOARD POT! Replacing the pot and reseting the zero ref discriminator level got the beast afoot again.

The lesson in this tale is that all problems can only be solved through first understanding the core issues and then the application aspects of the problem and finally attacking the major issue through hard work based on the aquistion of the understanding.

The fusor and its operation, the instrumentation, the vacuum systems are all part of this larger program.

If you want to get on board the hands-on fusion band wagon, a good base level understanding is needed. Reading a lot of good reference books is nice, but this forum will certainly take a lot of pain out of the startup due to free, hard won and sage advice given by its members.

P.S. The total was 22 hours on the warmup and 5 minutes for replacement of one pot.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Post Reply