NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/sci/electron/21265

Internet Message Format | 1992-12-21 | 5.4 KB

Path: sparky!uunet!usc!news.service.uci.edu!balboa.eng.uci.edu!gkwan
From: [e-mail] (Greg Kwan)
Subject: Re: Basic bench equipment query
Nntp-Posting-Host: balboa.eng.uci.edu
Message-ID: <[e-mail]>
Newsgroups: sci.electronics
Organization: University of California, Irvine
Lines: 116
Date: 21 Dec 92 08:40:25 GMT
References: <[e-mail]>

In article <[e-mail]> writes:
>
>I'm preparing to outfit myself with a basic bench setup, ie scope, function
>generator, power supply, DMM, and transistor tester (lots of in-circuit
>go/no go testing).

[...]

I was going to reply directly, but thought others may find something
useful, so...

>What's a more reasonable factor than 5X for applications where accuracy
>is less critical?

I don't know about that 5X factor. Bob Pease recommends 100MHz min BW;
Jim Williams says 150MHz, or 100MHz for a well qualified scope (one
that you know inside out). These are for industry, though. I'd say
20MHz is plenty for most hobbyists. Higher BW lets you see little
funny things (noise) riding on your nice signal, but the higher cost
probably isn't necessary for hobbyists. Also remember that 100+MHz BW
scopes have a handy 20MHz BW limit switch. Sometimes seeing that noise
isn't desirable.

For digital work, mostly you'll have a high frequency clock divided
down into reasonable frequencies. You really don't need to exactly
measure that HF clock, just know that it's there.

>Can I use a 20Mhz scope on waveforms that approach 20Mhz and get
>reasonable accuracy and visibilty?

Digital--yes. Analog--no. Reason--digital you're looking for presence/
absence of a clock, analog you need to accurately measure frequencies.
Note, too, the 20MHz range is usually w/ the horizontal multiplier (5
or 10X usually) on. The trace will be dimmer, and you'll really be
pushing the scope to its limits (of course--otherwise it'd be a 35MHz
scope :).

Unless you're intending to play with ~50ns single shots. Then I'd
say you need a 100+MHz digital storage scope.

>I definitely need dual trace, but how much do I need dual time base? Delay?

Probably not as critical. I've only used them a few times. I'd much
rather have quad trace and cursors (although those scopes will most
likely have all the features).

>Are Elenco, Hitachi and B+K Precision scopes in the <$1000 range reliable?

Sure. I have one of the generic (like Elenco) 20MHz scopes. 6+ years
old. I had an excellent Hitachi V-1150 (at work). The only complaint
was that its fan was a little noisy. A plus for the Elenco type
scopes is that they usually come with schematics and calibration
info.

>If not, I'd definitely go with thr Tektronics.

If you can afford it, Tek's are the real thing. My boss had a 2445.
It was like the V-1150 but could measure rise times and other stuff.
Great feel. Pity how the budget didn't figure one in for me. I did
like the V-1150's built in freq. counter, though.

>Tek has eliminated several lower-end scopes and the choices now are
>limited to the 20Mhz single-trace 2205 at $700 or the 60Mhz TAS455 for $1550.
>That's a hard choice, and I'm surprised there's no middle ground.

That is a bummer. If you're game, what about a nice used 465?
(I'm dreaming about a Fluke 97, though.)

>Any advice? Pointers?

Here's my opinions:

1. Get the best DMM you can--it's the most important test instrument.
I have a Fluke 77 and have been very happy with it, but would probably
go for an 87 or 79 now. You'll probably use the freq. counter since
it's a pain to fire up the scope just to measure a frequency. Capaci-
tance measurement is obviously quite useful. The analog bar graph does
come in handy--I use it often. Continuity/diode beeper--essential.
True RMS? Can be handy, but I don't consider it essential. On the
other hand, the freq. resp. of the meter is very important. The 77
is only good out to a few kHz, if I remember right. The 87 goes out
to 20kHz, not sure about the 79. (The 8060 is good to 200kHz.) An-
other good feature of the 87 (not sure if the 79 has it) is the
recording mode. Great for determining intermittent connections/
problems among other uses. I thought I was going to use the touch-
hold feature a lot, but it's kinda a pain to use (slow and not as
responsive as I'd like). You may find it useful, though.

2. Good soldering iron. I recommend a Weller WTCPS. ~$110+tips.
Well worth it. I have an Unger variable temperature system (~$150),
but find its heater too fragile. $15 a pop. I'm probably going to
buy a WTCPS...

3. +/- 0-20V, 1A (preferably tracking and metered) and +5V, 3A
power supply. Might be good to roll your own since you'll proba-
bly fry it a lot and should know how to fix it. $50-500

4. Function generator. I built my own (XR-2206 based) from a
Jameco kit. $15 + power supply (home grown again) and case, but
I've been thinking about buying a real one with a frequency
counter built in.

5. Scope. Tough call. I'd say either a basic 20MHz Elenco type
dealie or a Tek 465 if you have a little more and are somewhat
adventurous. Another followup recommended component testers.
My basic scope has this built in and it is actually quite use-
ful. The old lab manual for The Art of Electronics showed how
to build one and Hameg does sell one to add to any scope. Just
to add a little confusion.

6. Bob Pease's Troubleshooting Analog Circuits book. He includes
his list of essential tools. $35.

I dare you to ask me about audio test equipment. :-)

greg