|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
15th Jun 2020, 4:48 pm | #1 |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Hi-Z input probe by Bob Pease.
Hi-Z input probe by Bob Pease, the well-known and highly accomplished design engineer at National Semiconductor who sadly passed away in 1969. (Info. from Wikipedia).
The cct. diagram of this probe is attached. Note that there are two versions: the original, referred to here as 'V1' and a revised version, 'V2', which uses SOT FETs. There are three things about this probe that I do not understand, so enlightenment, please! 1. The claimed Rin is 1*e11 Ohms, which to me is 10*10 Ohms, i.e. 10^4 Meg Ohms. I cannot see how the Rin can be that high. 2. The claimed Cin is 0.23 pF. But the Cgs alone of Q1 will be approx. 5 pF. 3. What is the purpose of the 620 Ohm R at the gate of Q1? Al. |
15th Jun 2020, 5:03 pm | #2 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
I haven't seen that version before, I've only seen the original V1.
I'm nervous to comment on the design because the last time I reviewed the original version (elsewhere on the internet) I was slated for daring to point out any limitations of a Bob Pease design However, one major issue with the original was that it was an obvious generator of negative resistance (at HF and VHF) at the gate of the input JFET. I think the 620R series resistor has been added to try and combat this. I think the original version didn't use the shunt 30M resistance and this was optional. Without these resistors the input impedance at low frequencies will be very high indeed. With the Q2 JFET acting as a current source the voltage at the source of Q1 will track very closely with the signal voltage at the gate of Q1 and this means the effect of the Cgs capacitance will be effectively nulled as it has virtually no differential voltage across it.
__________________
Regards, Jeremy G0HZU |
15th Jun 2020, 5:08 pm | #3 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Hi-Z input probe by Bob Pease.
His Beetle was from'69; Bob died in 2011. Very sad, especially given the event he was driving home from.
While I've not read Bob's write-up of this, I'd say that the quoted input impedance must be that of the JFET itself. The input capacitance of the JFET follower will be bootstrapped, and the 620 ohm resistor is a "stopper" to prevent HF oscillation from low impedance sources. I've got a couple of his books - I'll take a look to see if this appears in any of them... |
15th Jun 2020, 5:12 pm | #4 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
I just spotted that this version uses a BFJ309 rather than the original 2N548x JFETs. I'm not sure the input capacitance will be as low as the original design.
I think the original design also relied on using a probe tip that was very skinny and it was arranged to give very low stray capacitance. I'm not sure what the input capacitance will be for the V2 version but I doubt it will be as low as 0.29pF.
__________________
Regards, Jeremy G0HZU |
15th Jun 2020, 5:18 pm | #5 | |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
Quote:
__________________
Regards, Jeremy G0HZU |
|
15th Jun 2020, 5:23 pm | #6 | |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Hi-Z input probe by Bob Pease.
Quote:
Available here: http://preview.kingborn.net/27000/4c...20212e7a4b.pdf |
|
15th Jun 2020, 5:31 pm | #7 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Hi-Z input probe by Bob Pease.
Quote:
Al. |
|
15th Jun 2020, 5:37 pm | #8 | |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Hi-Z input probe by Bob Pease.
Quote:
Al. |
|
15th Jun 2020, 5:50 pm | #9 | ||
Dekatron
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,924
|
Re: Hi-Z input probe by Bob Pease.
Quote:
B
__________________
Saturn V had 6 million pounds of fuel. It would take thirty thousand strong men to lift it an inch. |
||
15th Jun 2020, 5:56 pm | #10 | ||
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
Quote:
However, someone must have used the J309 in this probe without any obvious issues.
__________________
Regards, Jeremy G0HZU |
||
15th Jun 2020, 6:01 pm | #11 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: Hi-Z input probe by Bob Pease.
I'm interested, too- and I promise not to throw any brickbats! I find it very odd that someone would have got shirty about constructive criticism on any subject, but the 'net can be a funny place. Bob Pease may have been a benign sort of person to be a craven disciple of, but folk being unable to countenance infallibility of an idol has been a Very Bad Thing indeed throughout history
|
15th Jun 2020, 6:10 pm | #12 | |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
Quote:
One thing to be aware of will be the spread in Idss between JFETs. If it is assumed that Vg and Vs of Q1 are always at the same potential then I'd expect the JFET to run at Idss all the time. If the drain of Q1 is held at 15V and the gate of Q1 can be probed down to maybe -10V then this puts 25V across the JFET. If the Idss of Q1 is about 15mA then I'd expect to see this JFET get very hot when probing negative dc voltages. These voltages are also very near the limits on the datasheet. Therefore, I'd recommend using the 2N5485 rather than the 2N5486 and it might be worth an initial check of the Idss of any BFJ309 that gets fitted. It might get very hot when probing negative voltages if the FET has an Idss up near the max on the datasheet. At the minute all this is guesswork. If I get time this evening I'll have a go at simulating the probe.
__________________
Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 15th Jun 2020 at 6:16 pm. |
|
15th Jun 2020, 7:32 pm | #13 | |||
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: Hi-Z input probe by Bob Pease.
Quote:
Quote:
Quote:
Having said all that, let's now not drift off-topic, please. Al. Last edited by Skywave; 15th Jun 2020 at 7:40 pm. |
|||
15th Jun 2020, 7:48 pm | #14 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
The other area I recall being a bit confused about was the output connection that 'feeds to a scope'. I'm not sure what format this interface took in the original design. Was it by a very short piece of (RG174?) coax or was it with short flying wires?
Has anyone built the probe and experimented with different methods to interface to a scope?
__________________
Regards, Jeremy G0HZU |
15th Jun 2020, 9:49 pm | #15 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: Hi-Z input probe by Bob Pease.
Reading over the link in post #6, it's almost as if the active probe is being presented as "outline sketch", i.e. an illustration of topology with potential for development for specific roles- again, I see no problem in refinement and discussion around improvement in various aspects. It's not as if it's holy writ, more wise guidance.
Colin |
15th Jun 2020, 10:42 pm | #16 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
|
Re: Hi-Z input probe by Bob Pease.
Remember that Bob Pease was principally concerned with low frequency electronics, not RFery. Those designs are definitely not RF beasties and I've got doubts about those input capacitor figures.
David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
15th Jun 2020, 11:09 pm | #17 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
Agreed. I think the 0.29pF figure for the original circuit may be a bit optimistic.
The original version based on the 2N5485/6 will have some contribution from Cdg inside the JFET and also from the free space capacitance of the lengthy probe tips. I suspect that the capacitance will also vary slightly depending on the dc offset presented at the input at the gate of Q1. I think the J309 version (developed by someone else?) will have higher input capacitance because the process 92 JFETs have a higher Cdg capacitance. I suspect that this probe was developed for looking at high impedance 'instrumentation' electronics so it would rarely encounter any inductors in a circuit. I do think it will be prone to instability when probing inductors.
__________________
Regards, Jeremy G0HZU |
15th Jun 2020, 11:19 pm | #18 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
|
Re: Hi-Z input probe by Bob Pease.
It keeps reminding me of a problem I never found a good solution to, producing a 1.8k//1.5pF termination for a crystal filter I was using a BF904 dual gate mosfet in the following stage and that's about 3pF, so I was using 3.3 pf series capacitor and taking a hit on the voltage gain. It still didn't work because of pad track and stray C. It was a losing game. The other approach was a tuned circuit to get some negative capacitance in there, but component tolerances meant it needed adjusters and adjustments were most strongly not wanted. Bit of a dilemma.
david
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
16th Jun 2020, 12:06 am | #19 |
Dekatron
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 4,241
|
Re: Hi-Z input probe by Bob Pease.
Yes, the claimed BW is 90MHz, which is fast in my book, but nowhere near the dizzy heights of RF wizardry that you and Jeremy seem to enjoy
But no mention of BW requirements have been made by the OP in this thread yet, so I don't know if this is a follow-on from previous threads or just a general enquiry about the circuit as presented? One thing that isn't clear is how the input capacitance was determined. I'd like to think that Bob didn't just pluck that number out of thin air, or just loosely calculated it based on the known value of Cgs and the "gain" of the first transistor (a bit like PA speakers that quote their maximum SPL based on sensitivity and quoted power handling, which ignores a whole host of practicalities - not least thermal compression). So I would guess that he measured it in some empirical way - perhaps with added known series capacitance at the input, or perhaps a high value of series resistance instead. Or both, like a normaliser used for 'scope attenuator adjustments. He was also able to state the capacitance added by the FR4 probes, so he must have had some method for doing this. He was famously quite good at responding to correspondence, but sadly we're too late now. |
16th Jun 2020, 12:42 am | #20 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Hi-Z input probe by Bob Pease.
At a guess I would say that if the original 2N5485 JFETs were used then the input capacitance would/could be just under 1pF. This is based on the numbers in the datasheet and also on a basic simulation in Genesys. There may be some spread in this across a sample of 2N5485 JFETs but that really is just a guess. Even if a typical homebrew 'copy' only achieves 0.8 to 1pF it is still very low.
My greater concern would be the significant levels of negative resistance it can generate. The simulation predicts this occurs from the middle of the HF band up to about 250MHz. This won't matter much when probing high impedance circuits comprising opamps, resistors, capacitors and transistors but it can matter if there are inductors in the circuit.
__________________
Regards, Jeremy G0HZU |