- posted
8 years ago
-- John Larkin Highland Technology, Inc picosecond timing precision measurement
-- John Larkin Highland Technology, Inc picosecond timing precision measurement
Wow, and here a few weeks ago I was complaining about those $3 0805 resistors you were using.
Just goes to show that once you start downhill, it's hard to stop. ;)
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant
It is one GIG-Ohm, that has to be a bit tricky. Just breathe on it and it woun't be a gig Ohm anymore. I have no idea how well it does after SMT reflow and cleaning. Also, if you just buy 100 of them, they are ONLY $5!
Jon
I've designed circuits for customers with 50G SMT resistors. You need to put a slot under them to prevent water-soluble flux from destroying their performance. Sufficiently thick top copper combined with an unplated hole underneath can allow you to clean adequately between the resistor and the board, but a slot is better. Once you get it right, they're nice and reliable, but it doesn't happen by accident.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant
1G is easy.
Just don't use water-wash flux.
I have some samples of an 0805 1T resistor.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement
snip
I remember needing a 68M resistor for a timing circuit in my youth when I didn't know how to do it any better but what needs 1T? Isn't that the same as an open circuit? :)
I went to its catalog page and there was a 25 G ohm one! That's hard to get my head around. For example, wouldn't the air between the terminals be less (much less?) than 25G?
Bob
A clean PCB will have fA leakage. 25G is perfectly usable.
Here's an ugly rosin-flux-glopped, bare-finger-handled PCB.
Pins the needle on the 1e14 ohm range.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement
Den mandag den 12. oktober 2015 kl. 23.31.54 UTC+2 skrev John Larkin:
just be glad minimum order isn't a reel with 5000 ;)
-Lasse
I just did a current source with five 1 Meg's in series. Above ~1 Meg precision resistors disappear.
We pay about $3 for TH 1 G 1%. Big blue things, fat leads.
George H.
Pretty good for a 11kV SMT part. That's less than 100u$/V.
In LTspice you'd have to have 10 open circuits in series to get there. ;)
For TIAs at really low currents, you win by using a capacitor as the feedback element and fixing it up afterwards.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant
On Mon, 12 Oct 2015 18:00:48 -0400, Phil Hobbs Gave us:
snip
Yes. And even then, it is nearly impossible to get repeatable performance from them. Even an epoxy dipped planar types change value when environmental effects get applied.
In the HV lab, we used to call it "coffee breath".
Very difficult. We also used to place slots under our SMT HV diodes and caps on our miniature HVPS designs.
Still a quite susceptible realm at those form factors.
...and stay MILES away from "no clean".
** One app is externally polarised condenser microphones.
Two IG ohm resistors are needed, one to supply charge to the capsule and another to feed bias voltage to the gate of a FET or grid of a triode acting as the first preamp.
The large value allows the puny 20 to 100pF capacitance of the capsule to work down to 20Hz PLUS bypass and eliminate nearly all the resistor's thermal noise voltage.
... Phil
Your meter is broke ;)
Did you think of re-arranging the circuitry so only ONE resistor is needed (for both) functions?
They seem to charge by the ohm.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement
Cheaper but bigger, 1% tolerance, good to 1.5kV:
Cheaper but bigger and good to 3kV:
It's them thar Texans: ohm, ohm on the range...
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.