low price PC based Oscopes

[daniele]

Richard Hull wrote:
> However, no computer is fast enough for high level apps and boxes of all manner with specialized, hardwired analog and digital circuitry will have to be made up to accomodate the computers plodding and ever increasingly bloated operating systems.

> The computer will never be the heart of any cutting edge instrument, just the ass end that links to us meat engines. Likewise, computers just can't play where megavolt per meter fields and multimegawatt pulses abound. Can can you see plopping down a modern laptop to scope out a 16 megawatt 50kv pulse on a bench or scoping out a 10kw Tesla coil system's primary circuit. Even regular instruments fear to tread here.
...



Probably because we're at the beginning, dedicated hw is often the answer... and Windoze is not the only OS out there... 15 years ago the pc was not the same of today, hw was completely different and nobody would give a **** to it.
Me too was surprised in so much interest around that "box" full of CP/M


I agree with you Richard, infact I was talking about a part of the business...
Strong RF is deadly for a PC, but you teach me that oscopes are used in so many environments.
Maybe it will not be a cutting edge, but it will be in many labs.

The PC is absolutely NOT a solution for every problem, but it helps a lot in so many fields !

Daniele

[guest]

High end multi megavolt computer based oscopes will be out in less than 5 years.... They use remote fiber optic pickups... and pulse hardened computer main systems. But like all new crap out it's going to cost out the butt. 50,000 up. Uncle Sugar is buying the first batch... in about twenty years you'll probably see them on Ebay.

Larry Leins
Physics Teacher

[DaveC]

The typical PC based digital scope will be a fairly low frequency device. I would be surprised if this particular device had a bandwidth of more than a feww hundred kilhertz.

For comparison you can check out the National instruments or less costly Computer Boards data acquisition cards. They both work with Lab View, which has virtual instruments to simulate practically every commercially made instrument.

The PC cards are multichannel typically up to 16 single ended channels, or 8 channels in differential mode. But real time frequencies are again rather low.
I think most are well below 20 MHz bandwidth.

You can always do sampling and get wide "bandwidth" with a limited bandwidth system, ...provided.... the input amplifier (the vertical amplifier) has sufficient frequency response to actually detect the signal.

Most of the new DSO's Digital Sampling Oscilloscopes use sophisticated sampling techniques for most of the routine signal displays. The absolute number of samples per second, is the clue as to how fast they can actually record data. For example, the new TEK 1000 and 2000 series scopes record at 1 -2.5 Gigasamples per second. For a display consisting of 100 data points, this gives a realtime upper frequency of .... 1000 to 2500 divided by 100 or 10 to 25 MHz!! Yet they are rated at 60 to 200MHz. This is possible for repetitve signals. The real-time resolution or time interval between samples is 0.4 to 1.0 nS. So if a glitch or pulse is at least this time interval in duration, the scope can actually "see" it and record it as a line width pulse.

The older analog scopes, right now represent an enormous potential bargai for the home experimenter, since everyone in commercial businesses is focussed on acquiring the latest digital gadget.

In the last 2 years, I have bought a Tektronix 7834 analog storage scope (450 MHz real time bandwidth) with all four plug-ins including two matched 7B53 time bases that go to 0.5 nS per div sweep rate. With everything going, I can have 8 channels (4 each at two different sweep rates) on dthe display.

I also stumbled onto a batch of 7912AD's which are among the world's fastest Transient Digitizers, still . (100 Giga Samples /sec!!) But... they only handle a data stream of 512 points... Still at 10 pS per data point, not too much gets by. It captures 500 pS risetime transients with ease.. and displays them for as long as you want... at full brightness. Nice features. Plus... the entire instrument is computer controllable... provided you can speak TEK basic.



Costs? The TEK 7834 - I paid $800 for everything, original price... about $25K in the late 80's.

The 7912AD with monitor.. I got for $195. I had two of these in Lab at the Power Company, paid about $60K each for them. I wish now, I had bought the other 5 in the surplus store.

The newest TEK digital transient scope has a 6 GHz bandwidth, but still only does 20 Giga Samples/sec. It's a steal at only $57,000 plus options!!

So... my simple message is... don't over look the analog equipment. There are some absolutely wonderful instruments out there right now, available for pennies on the dollar.

Finally, for highest freq. reqponse...I think the TEK equipment holds the edge. But HP scopes are not far behind. And at one point, they had an exotic 70 GHz scope, but the price was way above $100K , and I think it needed Liq. Nitrogen for the front end.

Dave Cooper