Needing P6243 (or similar) 1GHz active probes
Yann Guidon / YGDES wrote 12/21/2016 at 19:44 • 0 pointsHi !
I finally got a Tektronix 11302A with 1GHz preamps (11A71 and 11A72) but no probe :-/ I can get away with the 400MHz "standard" inputs (11A32) for a while but I need probes like the P6243 to really get the most out of this crazy analog scope. And they are expensive. Multiply by 5 inputs and it's a very significant budget !
Any help is welcome :-)
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.
If your target circuit can handle a 1k loading or so, you can make Z0-probes that will do 1 GHz with a little care:
http://paulorenato.com/index.php/93
http://www.sigcon.com/Pubs/straight/probes.htm
I have been soldering 21x "probes" like this right on boards, and they work pretty well.
Are you sure? yes | no
Hey, that is an interesting "placeholder" until I find suitable probes :-P
I'm not sure my circuits will like 1K loading though. 100K probably.
Are you sure? yes | no
You could conceivably make a 2001x resistive probe the same way with a 100k resistor, but that's a *lot* of attenuation, and 100k resistors might have a spiral element and look inductive.
10k might be possible, though, for a 201x probe. You'd have to try it.
Are you sure? yes | no
Yep, it's going to be... fun...
Now if you have a similarly simple method for a really fast active probe...
I have seen one, maybe a year ago, but lost the bookmarks in a computer crash.
Why am I concerned by costs ? Because my setup can accomodate 5 1GHz inputs and they might become very useful for #Germanium ECL :-D
Are you sure? yes | no
I've wanted a fast active probe - and a fast differential probe - for a long time. No designs, here, though :-(
Maybe it's as easy as putting a modern high-speed op-amp on a wire :-)
Are you sure? yes | no
I don't think so :-D
I don't think I want to go beyond 10x because noise might create problems. The 11A72 amplifier can resolve 1mV/div but I suppose that's a practical limit, I think, and this sets the maximal ratio to 20x.
Given that I might have to resolve signals in the 100mV range, I'll have little headroom.
Are you sure? yes | no
Hopefully an active probe could help me remain at 1x. Maybe with a transistor follower circuit. I have a few AF280 (among others) to play with.
OTOH, reducing the current in the cable also reduces the skin effect and enhances the signal shape accuracy, which also explains why the "dumb series resistor" trick works well.
Are you sure? yes | no
There are some interesting links in this thread. All come with some trade-offs, but maybe it gives you an idea:
http://www.eevblog.com/forum/projects/good-design-for-a-diy-active-probe/
Are you sure? yes | no
I just tested the 11302.
At the 50 Ohms amplifiers (11A72), I have little headroom and I must avoid attenuation if I want to precisely "see" 100mV signals.
The more classic amplifiers (11A32) have more headroom but less bandwidth... but this is apparently not such an issue because the maximum horizontal resolution is 5ns/div :-/
Are you sure? yes | no
I know it's heresy, but I've always wanted to try an MMIC amplifier in this role. Maybe you can get enough gain to compensate for resistive probe loss. Sure, there are noise and distortion concerns, but MMICs cheap, high-bandwidth, and widely available.
Are you sure? yes | no
any link to this magic circuit ?...
Are you sure? yes | no