TEK Scope Probe question

I'd be wary of any claimed 400 MHz passive probe. Every pF is 400 ohms of circuit loading. 10.8 pF becomes a 37 ohm load at 400 MHz.

The only really fast passive probes are the resistive divider type that work into a 50 ohm scope.

No argument. That's what Tektronix calls there 400Mhz probe. I was actually looking for the P6136 probe that is a 350Mhz probe, but I saw this at a good price and it will adjust to peak with my scope.

So, Do you think these are real Tektronix probes or a Chinese knock off? Just curious, I think I'll buy them either way. Mikek

btw, here's a useful Tektronix probe specification list that goes up to probe model P6149A. Helps me, but could use an update.

a P6201 set worth anything to you? I've only used it a couple of times for my 485

Jamie

No thanks, 300MHz is way more than I need, but I'd just like to complete the scope. Mikek

For looks, one cannot tell; as a "knock-off", it is a perfect job. Verrrrrrrrrrrrrrrry remote possibility of being original Tek.

If they said "new" I'd be very suspicious. But, they say "used", and most scopes at lease/rent outfits had a full complement of probes included. I have some old 100MHz Tek probes that I've been using for DECADES, they are amazingly robust. The Probemaster probes that came with my new B&K scope in 1986 lasted, LITERALLY, 2 WEEKS under the same conditions. I'll never use anybody else's probes.

Jon

I agree. IMHO there is no use trying to measure anything over 100MHz with a high-Z probe.

That's perfectly true, however, consider this:

A 100MHz BW 'scope, with a 100MHz probe will only have an overall bandwidth of 70.7MHz, risetime degraded from 3.5ns to 4.95ns.

The same 'scope, with a 400MHz probe, will have an overall BW of 97MHz, and a risetime of 3.6 ns.

That's a 27% improvement in risetime.

Assuming, as always, a first-order rolloff.

I think the used part is a ruse, nobody would believe they could buy Two new Tektronix 400MHz probes for $145. Two used is still a bargain. Two Chinese well copied, still good. Two poorly made Chinese knockoffs, a waste of money. Decisions, decisions. I just searched 1000 deep in their reviews, Lots of happy 100MHz probe buyers, 1 happy Tek 500MHz probe buyer. No 400MHz buyers. Mikek

Is a 100 MHz probe down 3 dB at 100 MHz?

Didn't used to. Attenuator probes dramatically extend the LF response when AC-coupled, but had little effect on upper cutoff. At least, that's how it used to be with the old analog scopes from Tektronix and I'd guess HP. Tek had 50-ohm terminators with a probe coupling which made this easy to test. How specmanship games might be played these days makes this less certain.

Yes.

Is that not the definition, John?

How did you come to that staggering conclusion?

Cite reference.

How come my Chinese "100MHz" probe still goes out to >400MHz?

I didn't say the phase or amplitude was anywhere near correct... it's probably within 3dB most of that range. It's just a bunch of wires and resistors and stuff, nothing fancy.

But that's the real problem, getting clean step response: low phase shift and fractional dB amplitude out to frequencies much higher than rated. That's where the microstrip and layout and funny business gets involved in probe design.

Tim

What's its quoted risetime?

What does a fast edge look like?

Beats me!

Says Aidetek P6100 on the BNC end.

Standard grounding clip, making no attempt to neaten its path:

Top = coax cable output from the circuit (call this the reference), bottom = probe (no, I don't have a manual 10x option on the scope, so it appears to read low by 10x). Measuring the same node in circuit. (Obviously, the probe cable is shorter by a few ns, safe to disregard the time difference. Also looks that the cable isn't quite matched or terminated well on one end or the other, judging by that little blip 25ns later.)

Coaxial tip grounded to board as best I can manage:

The probe itself looks like it's got a little peaking around ~100MHz, making the pulse overly sharp on the trailing side. The really sharp stuff doesn't appear to be appreciably slowed down, though its amplitude probably isn't real great (it seems to be 2.5Vp-p rather than 2V, or

What'll really cook your noodle is this: the sharp pulse *still* comes through with the ground clip removed altogether. Lower frequency stuff is where it gets screwed up -- 10s of MHz, I think corresponding to cable length 1/4 wave, 3/4, etc. resonances I think. In other words, there's a huge drooling, bouncing residual following the pulse.

When all you want is an idea of what's going on, your probing can be really awful. It's when you need to squint at it and take measurements say 50% or better that you need to get off your ass and prepare something good enough to be useful (a proper FET probe, or a coaxial probe connector, or a 450+50 ohm divider into coax, or even a proper buffer straight to 50 ohms).

Tim

I don't want to get into an argument over mathematical equivalence between impulse and step response...especially when the impulse really isn't...

I find that a response to a well-behaved step is far easier to interpret in the context of actual scope applications.

What, you can't integrate in your head? :-)

Of course, it's not an impulse, either... it's a short pulse. And not even gaussian at that. Some horrible ugly exponentially tailed hyperbolic whoknowswhat. :)

At first I was going to do a rectangular pulse, but this particular circuit is better made for short pulses. The slower edges, extra ringing and soft top cluttered things up more. That or the pulse line (cable) I plugged in really stinks.

Tim