Another lame "which probe?" thread

[KarlMonster]

I did search and poke around as best I could, but I didn't quite see a thread that satisfied. If there is an old non-stuck thread that covers this, please point me there.

I have a recent AAS in Electronics and Computer Technology (plus other degrees, yada yada) so I have some idea what I'm doing. I didn't know that much about scopes 3 weeks ago, but I've learned a lot and understand most of it. I've recently acquired a Tek 2336 100MHz (dual trace?) scope, and am looking for probes to go along with it. The manual lists part number 010 6108 03 as the appropriate probe, which I'm guessing would be the Tek P6108A 03 probe (right?). [Whoops, new information says I'm wrong, its the P6108 with no suffix.] According to Tek, there is no modern analogue for that discontinued probe. The same Tek guide says the suggested replacement is P6105A - also discontinued with no modern analogue.

So, a few questions:
Should I try to match the specs (13pF input C) of the original probe? Will any quality 10x probe do - so long as it compensates for the scope's 20pF input?

Should I defer to Tek models?

(The part you've been dreading) Is there a recommended list for probes?

Can anyone point me to the equation for system (scope + probe) bandwidth?
I understand its different for passive and active probes.

Thanks for your time.


Edit: Through sheer dumb luck I've located a (010 6108 03) P6108 probe (eBay). Pictures show it with and without the 10x wire on the BNC, so I assume I'll get the 'without' version.
However, I reckon there's a chance of it being a DOA probe, so my questions stand.

[dfmischler]

What are you trying to look at?  The low-cost import probes (AKA "China's Cheapest") available on ebay are adequate for a lot of purposes, and really cheap.  When they no longer suit your purposes you will probably know, or soon thereafter.

[KarlMonster]

I may have confused myself by trying to search the forum first. I found this post on probes for the Tek TDS 754A which I interpreted as "use probes intended for the 'scope". However, I didn't look up the scope number to see that the model is a mere 18 years old (based on when Tek says the manual was first available). Probes for such a young 'scope are prolly not hard to find. Then just now I found this thread which was more to the point: "use any old probe" (that compensates appropriately). So yeah, I'll go look for a 10x that includes a feature to ring the butler to come and compensate the probe for me. 

To answer your question, my current project is tinkering around with low-pass RLC filters. Definitely low-voltage and low-frequency work since it involves a 1v peak-to peak signal at 240 Hz or less - though there should be a gain of about 35 later in the circuit. Additionally I have a cassette deck that needs attention, and I'm determined to fix that myself. Also Dave's "1x probe" blog gave me an idea to construct an active 1x probe using a D-MOSFET, as my RLC project makes use of one for the same purpose; passing a signal through the gate while preventing any signal contamination from 'downstream'.

At any rate, I'd still like to see the eq. for "system bandwidth", I'd be obliged if anyone could provide that.

[dfmischler]

You bought a 100MHz scope, and the cheap import 100MHz probes will be OK for almost everything.  A Tek TDS754A is a 500MHz scope so it needs lots better probes to allow it to reach its potential, so to speak.  Even the P6139A probes won't necessarily solve all the problems of seeing things at 500MHz because the probe capacitance becomes a big issue at higher frequencies.

[KarlMonster]

Would you believe that I found a PDF titled "System Bandwidth Calculations" that included no equations at all?

According to a Tek document that I tracked down, system bandwidth for the Tek TDS 794D can be approximated this way:

system risetime = sqrt( (scope tr)^2 + (probe tr)^2 )

system bandwidth = 0.4/system risetime

I'm assuming that 0.4 is a truncated 0.35, but it could also be that 0.4 is the correct factor for the TDS 794D. The manual for my Tek doesn't get anywhere near this level of detail, so the 0.4 is probably the more reasonable of the two. I may defer to the 0.4 as my unit is much older. And so is my oscilloscope. 

Now, when I come across a probe that is stamped with its tr, but not its B/W, I can determine (generally!) what effect that has on the system bandwidth. Such an enigmatic (and reasonably cheap) probe has appeared, marked as 1.4 ns tr, but no stated B/W. Using the 0.35 factor, that works out to 250 MHz - but I found identical probes listed as 80MHz (but with no visible basis for claiming that). The above tr calculations result in a system B/W of 92 MHz (assuming my scope performs at the rated 100 Mhz). Now I'm keen to get access to a good sig gen to see what the real system bandwidth is.

Does it make any difference for me? Not really, but I always like to see the numbers.
Its better when I understand where they came from.

Anyone happen to know the tr ratings of the "Eastern" 10x 100MHz probes on eBay?

Edit: Nevermind, I dug up two or three separate threads on those probe's quality. Why are they *all* 1x/10x ??