Topics

Which probe is better suited for 2465B?


Giuseppe Marullo
 

Got the scope without probes,

Is getting a couple of P6137 the right ones to be used as a general purpose
probes?



What about HF/VHF/UHF 50Ohm signals? Could I tap these with a dummy load or
do I need 50 Ohm specific probes/cables?



I also need to measure about 900-1100VDC and I was thinking to use el-cheapo
P4100 Probes. Any recommendation about it?




TIA



Giuseppe Marullo

IW2JWW - JN45RQ


ppppenguin
 

Hi Giuseppe

For probing 50 ohm signals it's best to use a simple passive low resistance probe. This also works well for fast logic signals. This will outperform all standard probes. Links to a couple of designs here: https://www.eevblog.com/forum/projects/lo-z-probe/

It's just a resistor, a length of 50 ohm co-ax and terminate your scope with 50 ohms. On a 2465B there's a switch for this. On lesser instruments you'll need a T and termination.

For high voltage you need a probe that is rated for that voltage.The P4100 looks suitable. Just take care around high voltages. If the high voltage comes from a low impedance source, such as a transformer direct from mains/line then it can kill you. I survived the experience but it hurt. A lot.

Jeffrey


Jean-Paul
 

Hello: Yes P6137 is the specified for the BW and Cin. Hard to find good one, many junkers out there.

For 50 Ohms use a coax like RG/198/U on a BNC.

For probing wideband we use a Zo probe P6156 eg 1K Ohm resistance but very wideband.

At 1 kV for HV we use the old P6015, the Freon filled units are all degassed but still can work to 10 kV. Available on epay ~ $50 but check if all parts present.

General info is in the Tek Circuit Concepts book, Oscilloscope Probes, highly recommended.

https://www.davmar.org/TE/TekConcepts/TekProbeCircuits.pdf

Enjoy,

Jon


Tom Gardner
 

On 23/02/21 14:34, ppppenguin wrote:
For probing 50 ohm signals it's best to use a simple passive low resistance probe. This also works well for fast logic signals. This will outperform all standard probes. Links to a couple of designs here: https://www.eevblog.com/forum/projects/lo-z-probe/

It's just a resistor, a length of 50 ohm co-ax and terminate your scope with 50 ohms. On a 2465B there's a switch for this. On lesser instruments you'll need a T and termination.
Agreed, but realise the termination is really 50ohms//15pF. Standard transmission line theory shows that will result in frequency-domain notches, where the frequency is determined by the cable length.

That's why I like my 485: it has a /real/ 50ohm attenuator! (i.e. not 1Mohm//15pF//50ohm :) )


Jean-Paul
 

Tom is the 485 input 50 Ohm free of any capacitance? Or just lower than the others?

In case the 15 pF is an issue a Zo proble or active FET probe brings the tip C to ~ 1 pF.

Jon


Tom Lee
 

Hi Jon,

It's certainly much lower in capacitance in the 50 ohm mode than in the 1Meg mode, but I've never actually measured it.

The official spec from Tek only gives SWR values for the 50 ohm mode: < 1.15 for all ranges except the 5mV/div setting (< 1.25 there). If one attributes all of the SWR to a shunt capacitance, the upper bound on that capacitance works out to be roughly 5pF with these numbers, assuming that I didn't punch the buttons on my calculator incorrectly. That number feels about right, in any case.

Given Tek's policy of sandbagging, my guess is that the actual capacitance is probably around 3-4pF.

-- Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 2/23/2021 08:56, Jean-Paul wrote:
Tom is the 485 input 50 Ohm free of any capacitance? Or just lower than the others?

In case the 15 pF is an issue a Zo proble or active FET probe brings the tip C to ~ 1 pF.

Jon




Tom Lee
 

I should add that I don't fully understand the reason for your question, though I was happy to answer it. The 485 is a 350MHz scope, and so one would naturally expect the input RC time constant to be small enough not to get in the way of the bandwidth spec in 50 ohm mode. With my 5pF estimate, one sees a time constant of 125 ps, or over 1GHz bandwidth at the connector. So all the numbers fit together sensibly.

Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 2/23/2021 08:56, Jean-Paul wrote:
Tom is the 485 input 50 Ohm free of any capacitance? Or just lower than the others?

In case the 15 pF is an issue a Zo proble or active FET probe brings the tip C to ~ 1 pF.

Jon




Jean-Paul
 

Rebonjour, the 485 spec is 50 Ω /1 MΩ//20 pF

https://w140.com/tekwiki/wiki/485

I assume the 20 pF applies to either Zo.

Best measurement of Cin is the classic TEK 130 LC meter, very easy to use and accurate, with a 1, 3 pF scale.

We also use the HP 4232A LCR meter.

Very interested to hear the 485 Cin actual data. I bet its >0 (:-:)

Bon Journee

Jon


Giuseppe Marullo
 

Hi John,

Hello: Yes P6137 is the specified for the BW and Cin. Hard to find good one, many junkers out there.
Trying to get them from Ebay, I asked the seller specific conditions/photos and I will claim the money back if they are not in good order. Anything specific to ask to be checked, like the button or something similar?

For 50 Ohms use a coax like RG/198/U on a BNC.
Ok, this is easy...but what about high power stuff, like 100W RF or similar? Probably a T plus a dummy load but then high impedance probe, right?

For probing wideband we use a Zo probe P6156 eg 1K Ohm resistance but very wideband.
Twice the price of a used P6137. Will have to wait a little.

At 1 kV for HV we use the old P6015, the Freon filled units are all degassed but still can work to 10 kV. Available on epay ~ $50 but check if all parts present.
I checked but prices are way different at the moment. I bought the P4100...like 10EUR shipped. How I will not die using it.


General info is in the Tek Circuit Concepts book, Oscilloscope Probes, highly recommended.
https://www.davmar.org/TE/TekConcepts/TekProbeCircuits.pdf
Great reading, thanks!


Enjoy,

Jon


Tom Lee
 

Non, non et non.

I gave you the upper bound on the input capacitance implied by Tek's input VSWR spec. That number is 5pF. It's irritating that I do this work for you and you simply ignore it. That diminishes my willingness to help you going forward, as it would be a waste of time.

It should be obvious that the 20pF number cannot possibly apply to 50 ohm mode. Do a calculation as I did. If you find an error in my button-pushing, let me know. Otherwise, accept the implications of my calculation.
--

Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 2/23/2021 10:41, Jean-Paul wrote:
Rebonjour, the 485 spec is 50 Ω /1 MΩ//20 pF

https://w140.com/tekwiki/wiki/485

I assume the 20 pF applies to either Zo.

Best measurement of Cin is the classic TEK 130 LC meter, very easy to use and accurate, with a 1, 3 pF scale.

We also use the HP 4232A LCR meter.

Very interested to hear the 485 Cin actual data. I bet its >0 (:-:)

Bon Journee

Jon




Giuseppe Marullo
 

Hi Jeffrey,
For probing 50 ohm signals it's best to use a simple passive low resistance probe. This also works well for fast logic signals. This will outperform all standard probes. Links to a couple of designs here: https://www.eevblog.com/forum/projects/lo-z->probe/
First link inside EEVBlog is dead, but the second seems really good:
http://www.emcesd.com/1ghzprob.htm


For high voltage you need a probe that is rated for that voltage.The P4100 looks suitable. Just take care around high voltages. If the high voltage comes from a low impedance source, such as a transformer direct from mains/line then it can kill you. I >survived the experience but it hurt. A lot.
Already bought, hope it will work thanks.

Giuseppe Marullo
IW2JWW - JN45RQ


Jim Ford
 

Well, 20 pF is about 22.7 ohms at 350 MHz, so that would affect the 50 ohm input. So I assume Tek left the 20 pF shunt cap off the 50 ohm input.

Jim Ford

------ Original Message ------
From: "Jean-Paul" <jonpaul@ix.netcom.com>
To: TekScopes@groups.io
Sent: 2/23/2021 10:41:46 AM
Subject: Re: [TekScopes] Which probe is better suited for 2465B?

Rebonjour, the 485 spec is 50 Ω /1 MΩ//20 pF

https://w140.com/tekwiki/wiki/485

I assume the 20 pF applies to either Zo.

Best measurement of Cin is the classic TEK 130 LC meter, very easy to use and accurate, with a 1, 3 pF scale.

We also use the HP 4232A LCR meter.

Very interested to hear the 485 Cin actual data. I bet its >0 (:-:)

Bon Journee

Jon





Tom Lee
 

Hi Jim,

Lower frequency scopes merely throw a 50 ohm terminator across the input to provide a 50 ohm mode, but the 485 has two separate attenuator sections, one for 1M, and a different one for 50 ohms, with a relay switching between them, as necessary. Because the entire 50 ohm signal path is different, the capacitance can be (and is) designed to be inherently lower, as it has to be to make the bandwidth spec.

Cheers
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 2/23/2021 11:00, Jim Ford wrote:
Well, 20 pF is about 22.7 ohms at 350 MHz, so that would affect the 50 ohm input.  So I assume Tek left the 20 pF shunt cap off the 50 ohm input.

Jim Ford

------ Original Message ------
From: "Jean-Paul" <jonpaul@ix.netcom.com>
To: TekScopes@groups.io
Sent: 2/23/2021 10:41:46 AM
Subject: Re: [TekScopes] Which probe is better suited for 2465B?

Rebonjour,  the 485 spec is 50 Ω /1 MΩ//20 pF

https://w140.com/tekwiki/wiki/485

I assume the 20 pF applies to either Zo.

Best measurement of Cin is the classic TEK 130 LC meter, very easy to use and accurate, with a 1, 3 pF scale.

We also use the HP 4232A LCR meter.

Very interested to hear the 485 Cin actual data. I bet its >0 (:-:)

Bon Journee

Jon








Tom Gardner
 

On 23/02/21 18:56, Giuseppe Marullo wrote:

For 50 Ohms use a coax like RG/198/U on a BNC.
Ok, this is easy...but what about high power stuff, like 100W RF or similar? Probably a T plus a dummy load but then high impedance probe, right?
A high power through attenuator (probably with cooling fins!), or a 20dB coupler which taps off 1% of the power going through it.


Tom Gardner
 

On 23/02/21 16:56, Jean-Paul wrote:
Tom is the 485 input 50 Ohm free of any capacitance? Or just lower than the others?
The 485 has two attenuators, the normal 1Mohm attenuator used with traditional "high impedance" probes, and a separate 50ohm attenuator. A relay connects the input from the BNC socket to either one or the other attenuator.

In the scanned service manual available at the usual places, have a look as the /two/ diagrams <1> for the "Ch1 attenuators (Hi Z)" and "Ch1 attenuators (Lo Z)". The relay is K151.

Other scopes' 50ohm input is merely a 50ohm resistor connected across the "Hi Z" attenuator with 1Mohm/15pF input.


Jean-Paul
 

Sr. Marullo,

Keep scope inputs away from TX or exciter outputs, there are various techniques to monitor the modulatiion, power etc. using directional couples and splitters.

A power rated load may have a tap off, but a resistive attenuator could be built, say 100:1

Take care of the max and peak power rating of the scope at 50 Ohm.

The broadcast engineers among us can advise better,

Ciao,

Jon


Jim Ford
 

Totally makes sense.  I was refuting Jean-Paul's assumption that the 20 pF was across the 50 ohm input.  We agree.  Thanks, Tom.              JimSent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------From: Tom Lee <tomlee@ee.stanford.edu> Date: 2/23/21 11:09 AM (GMT-08:00) To: TekScopes@groups.io Subject: Re: [TekScopes] Which probe is better suited for 2465B? Hi Jim,Lower frequency scopes merely throw a 50 ohm terminator across the input to provide a 50 ohm mode, but the 485 has two separate attenuator sections, one for 1M, and a different one for 50 ohms, with a relay switching between them, as necessary. Because the entire 50 ohm signal path is different, the capacitance can be (and is) designed to be inherently lower, as it has to be to make the bandwidth spec.CheersTom-- Prof. Thomas H. LeeAllen Ctr., Rm. 205350 Jane Stanford WayStanford UniversityStanford, CA 94305-4070http://www-smirc.stanford.eduOn 2/23/2021 11:00, Jim Ford wrote:> Well, 20 pF is about 22.7 ohms at 350 MHz, so that would affect the 50 > ohm input.  So I assume Tek left the 20 pF shunt cap off the 50 ohm > input.>> Jim Ford>> ------ Original Message ------> From: "Jean-Paul" <jonpaul@ix.netcom.com>> To: TekScopes@groups.io> Sent: 2/23/2021 10:41:46 AM> Subject: Re: [TekScopes] Which probe is better suited for 2465B?>>> Rebonjour,  the 485 spec is 50 Ω /1 MΩ//20 pF>>>> https://w140.com/tekwiki/wiki/485>>>> I assume the 20 pF applies to either Zo.>>>> Best measurement of Cin is the classic TEK 130 LC meter, very easy to >> use and accurate, with a 1, 3 pF scale.>>>> We also use the HP 4232A LCR meter.>>>> Very interested to hear the 485 Cin actual data. I bet its >0 (:-:)>>>> Bon Journee>>>> Jon>>>>>>>>>>>>>> >>


Roy Thistle
 

On Tue, Feb 23, 2021 at 10:56 AM, Giuseppe Marullo wrote:


Trying to get them from Ebay, I asked the seller specific conditions/photos
and I will claim the money back if they are not in good order. Anything
specific to ask to be checked, like the button or something similar?
I
If one searches the messages, for P6137, here on TekScopes... there is some discussion about these P6137 (and the other probes, in that family of probes.)
Those threads have some good discussion, about getting them from Ebay.
P6137 are known to be fragile.
Many of these on Ebay have obvious physical damage ( broken tip etc.)... but, some will also have damaged coax (common enough, and not obvious) That might make the probe intermittent... or cause it to become intermittent, with handling.
Unless you really need the P6137's bandwidth... older, lower bandwidth, more hardy, and less costly Tek probes... that can be compensated for the 2465B input capacitance... might server you better.


Giuseppe Marullo
 

Unless you really need the P6137's bandwidth... older, lower bandwidth, more hardy, and less costly Tek probes... that can be compensated for the 2465B input capacitance... might server you better.
That was the point, going past the performance offered by el cheapo Rigol 1054z (yes, already, hem, "upgraded") with its vanilla probes.

AFAIK there is not really a cheaper substitute for that probe, but this is new territory for me.
I only have a Tek since few days, the only other Tektronix hw I own is a couple of old Sony Tektronix LA), so I may be wrong and asking here.

I mean they are the standard match for the 2465B. I will play with DIY (maybe a 3d printer could help with the enclosure) but I would like to see more and better than what a 100MHz scope could do.

I also need the same for a 485 going to be delivered in the next days.


Jean-Paul
 

Sr Giuseppe

1/ The Chinese knockoffs probes are not capable of the specified BW or RT and will proove clumsy and unreliable.

2/ We have used P6137 for decades, they must be treated with care like any probe, use care kinking the cable, dropping or exerting pressure on the tip. They are robust and professional and far less delicate than some other TEK probes.

3/ On epay beware of probes without tips, damaged inner tip, frayed or torn coax.

Sellers that are equipment dealers will sell lots of 2-10 junkers. The tips and ground leads are needed and to trivial to find.

4/ Besides epay, we have found good probes at Ham Radio flea markets and silent key estate sales.

5/ Occasionally you find NIB probes with original accessories and case. Those are costly but the best.

Hope this is of use to you!
Bon Chance,

Jon