Pulse generator

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
Tom Dressel
Posts: 287
Joined: Fri Jun 22, 2001 4:44 pm
Real name:

Pulse generator

Post by Tom Dressel »

I had a picture of the scope tracing from my primative pulse generator designed to test the NIM components. Richard Hull said that a good clean pulse generator is hard to build. I an starting to believe that it is impossible to build.

I was using a 555 timing chip to generate a train of square wave pulses then a 741 op amp to differentiate the squaer wave and generate a series op spikes. The problem is that the 555 chip puts out a bunch of high frequency noise in the DC power supply, and I just can't seem to get rid of it. By the time the signal is attenuated down to the milivolt level the noise level is horrendous. As a last resort I will try an optical isolater in the last stage of the pulse generator drven by a seperate power supply.

Any other ideas?

Tom Dressel
guest

Re: Pulse generator

Post by guest »

A 741 is a terrible choice for a buffer. The slew rate is pretty slow. I'd take your 555 and feed it through a fast CMOS gate. Try a 74HC(T)04 or a 74FCT04. You'll need really good decoupling on the buffer, and you might want to actuall use gates from two packages in series, with separate decoupling on each, if HF noise on the power is a problem.

How are you attenuating it? Are you getting noise in through the ground? Are you sure the noise is really there, or is it being picked up on the leads to the oscilloscope?

Another approach is to use a fast transistor or fet as a buffer.
Tom Dressel
Posts: 287
Joined: Fri Jun 22, 2001 4:44 pm
Real name:

Re: Pulse generator

Post by Tom Dressel »

Sounds good to me. You have to remember that I am not an EE. How about posting a schematic for a device that works.

Tom Dressel
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: Pulse generator

Post by DaveC »

Tom - I do agree with Jim Lux... the 741 is too slow for your purposes. But since the 555 puts out several volt square waves, why not just feed the output through a capacitor in series with either a 50 or 75 ohm resistor to ground. The take your output from across the resistor and use a voltage divider to get the level you need.
Regarding noise.... check that your DC Pwr supply has high frequency caps on the output side. Especially necessary if its a bipolar supply (+/- voltages), since the negative linear regulators are notorious for self oscillation. But...... is your power supply possilby a switcher? If so.. that's likely to be the source of your noise.

One solution to this is to build a good filter for the supply output...needs an inductor in series and a capactor to gnd. This will block any current pulses from your 555 from getting into the supply.

Use your oscilloscope set to AC coupling and look at your DC PS with and without the 555 circuit load. If all is well, it is not unreasonable to expect the output to be clean at least to 1.0 mv with 5 to 10 volts DC set.

Dave Cooper
Tom Dressel
Posts: 287
Joined: Fri Jun 22, 2001 4:44 pm
Real name:

Re: Pulse generator

Post by Tom Dressel »

Thanks for the advice.

I am using a 7815 voltage regulator, with a 0.1 uf cap across the output to ground, to drive the 555 chip. (15 volts to ground). The op amp is driven by the 7815 and a 7915, voltage regulator again with a filter cap on the output (+15 volts to -15 volts). Possably one or both of the filter caps are bad.

Anyway, I think for starters, I will use the square wave calabration from the scope to drive the cap-resistor combo as you have suggested. I am not trying to get fancy I just need some fast pulses to set up the NIM components.

Tom Dressel
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: Pulse generator

Post by DaveC »

Tom -
I think using the calibration signal will be pretty simple to use. Depending on the age of your scope, the rise time will either be TTL compatible (50 nS rise time) or slower. The RC, (cap input, resistor to gnd) circuit forms a nice, simple differentiator, giving a pulse width approximately equal to the risetime of the square wave.


Regarding your present circuit, the 741 is a bit too slow, as we mentioned before, but the caps on your 7815 and 7915 's could be okay, just need some more capacity. The 7915 needs a 1 mfd solid Tantalum on the output, in parallel with the typically 10 or more ufd output electrolytics to keep it stable. I usually do the same thing for the positive regulator, too.

Wiil be interested to hear how it goes.

Dave Cooper
guest

Re: Pulse generator

Post by guest »

Bipolar 555s are notorious for injecting trash into supply lines due to cross conduction in the output driver stage. Any of the CMOS 555s will be much better in this regard. The later CMOS 555s such as the TI TLC555 and the National LMC555 also have much faster edges than the bipolar device. Radio Shack may even have the TLC555. If not, it can be found easily from a supplier on the Web.
Follow the output of the TLC555 with a RC differentiator, then use a capacitor coupled emitter follower circuit to buffer the differentiator. This can then be coupled to an attenuator, and followed by another buffer. A simple bipolar emitter follower will beat the pants off of a 741 for speed any day of the week. If you are willing to deal with just one polarity of pulse, the resulting circuitry can be pretty simple, and capable of being powered by a 9V battery.
Tom Dressel
Posts: 287
Joined: Fri Jun 22, 2001 4:44 pm
Real name:

Re: Pulse generator

Post by Tom Dressel »

Thanks again.

Everything I know about electronics I got from Forrest Mins handbook on IC chips.

I should be able to try some of the suggested fixes this weekend.

Tom Dressel
Tom Dressel
Posts: 287
Joined: Fri Jun 22, 2001 4:44 pm
Real name:

Re: Pulse generator

Post by Tom Dressel »

I picked up a low power TLC555 timer chip today, which should help with the noise problem. But I also found a Tectronix type 115 pulse generator at a salvage place for $20.00, (See IDJ post), which definitely solves the noise problem. Attached is a tracing of a clean 50 nsec. pulse.

Tom Dressel
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Pulse generator

Post by Richard Hull »

SURPLUS IS THE WAY TO GO. I have always said this and hamfests and surplus outlets can supply material and devices that we could not dream of replicating with any ease or functionality.

Sometimes we are fored to cobble something up due to time constraints or due to the fact that it just doesn't exist in nature. Pulse generators are not one of those items.

Tektronix, HP, GenRad (General Radio) and IEC (Industrial Electronics Corp) are among the best pulse generators I have seen and used.

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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Post Reply

Return to “Neutrons, Radiation, and Detection (& FAQs)”