2400 series probes (is p6137 the most optimal?)


B. Levin
 

I need a few more probes for my new 2465b.

I understand that the 6137 is the recommended probe that came with the scope.  however, I see plenty of 38 and 39 probes for sale on the bay.  is there any reason not to buy these instead of the 37, if I find one at the right price?  is the ID button any different?


ditter2
 

Yes, it does matter if you want to take advantage of the maximum bandwidth.

Tek passive probes over about 200 MHz are designed for a specific series of scope.  The intermediate time constants in the compensation network have elements that are tuned to compensate for the parasitic elements in the scope attenuator system.  These are mostly associated with the physical construction of the attenuator, including the relays, and elements on the hybrid circuit.  The tend to be very consistent from scope to scope of the same model, but because different designs are used in different models, they change considerably between models.

These effects minimize the overshoot and "wrinkles" in the first 2 ns of the rising edge and flat top.  Minor differences will only result in a "bump" where it is not suppose to be.  Larger differences will result in considerable overshoot or slow "bleedup" on the front corner of the pulse.  (for instance, don't consider using the P6139A on a 2465 - there us a huge difference)

If you don't care about the high frequency performance, then any probe will work assuming the LF comp range is sufficient.  But if this is the case, why did you spend the extra money on a 2465B?

- Steve


B. Levin
 

wow, great reply, steve.  I was not aware the little wiggles would be from this, but I guess it makes sense.

I want as accurate a waveform as I can get.  I am going to be making decisions based on it, so 'fidelity' does count, here.

ok, great - I'll stick with the 6137 and learn to be happy with the low selection on the bay.

let me ask one more: if I wanted to spend more and go active, what is the recommended way for that?  it seems there is a power supply box that is $300 (at the very least, used) that you need to use and that powers the active probes.  the probes seem to 'talk' to this box and that link is critical (for comms and waveform) but once the box is done, is the rest of the link all that important (going to the scope)?  do I have to care, now, about this box matching well with my 2400 series?  maybe its just not recommended?

thanks again!
/bryan


froggiegremlin
 

Hello Bryan: I have a LOT of TEK probes and several 2465, and a 2467B. 

The P6138 and P6139,are are much more delicate than P6137, and prone to damage. Even the ground clip is not as good but its mainly the probe tip and body that is the issue.

The 6137 is the most robust.

Its also very accurate and versatile, I have compensated for far beyond the specified compensation range, even hi cap inputs like old 7A26 TEK 7000 plugin, with 20 pF, 1M!.

Finally,  if you can stand the loading, use a Zo probe, which runs into a 50 Ohm input and have a loading impedance or 500 or 5K Ohms but MUCH lower capacitance.  See the TEK circuit manuals for oscilloscope probes for more on Zo probes. I have 2 of  the Zo probe P6156, these come in options with a switchable x1, x10, x20 znd x100, so buyer beware!

Enjoy!

Jon Paul
Crypto Museum


 

One thing I noticed about third party 1 MOhm passive probes faster than 100 MHz
is that besides the compensation adjustment, they include (or should include) a
whole set of transient response adjustments. During tests with my 2440, the
response of the general purpose replacements probes I bought for it was
indistinguishable from the 50 ohm terminated response.

Of course calibrating the probe and oscilloscope combination is not trivial
since it requires a flat level pulse generator with a fast enough transition
time and a coaxial attachment to the probe.

On 28 Feb 2014 16:02:46 -0800, you wrote:

Yes, it does matter if you want to take advantage of the maximum bandwidth.

Tek passive probes over about 200 MHz are designed for a specific series of scope. The intermediate time constants in the compensation network have elements that are tuned to compensate for the parasitic elements in the scope attenuator system. These are mostly associated with the physical construction of the attenuator, including the relays, and elements on the hybrid circuit. The tend to be very consistent from scope to scope of the same model, but because different designs are used in different models, they change considerably between models.

These effects minimize the overshoot and "wrinkles" in the first 2 ns of the rising edge and flat top. Minor differences will only result in a "bump" where it is not suppose to be. Larger differences will result in considerable overshoot or slow "bleedup" on the front corner of the pulse. (for instance, don't consider using the P6139A on a 2465 - there us a huge difference)

If you don't care about the high frequency performance, then any probe will work assuming the LF comp range is sufficient. But if this is the case, why did you spend the extra money on a 2465B?

- Steve


B. Levin
 

on the bay, I've seen 2 different bodies of p6137 probes.  one has no holes on the back of the flat(ish) comp box and the other has a series of scattered holes.

what is the difference?  I know there is HF comp inside the box as 3 adjustments and you have to remove the box cover to get to them.  the normal screw hole is the customary LF comp.

but what's the deal with the holes on the back of the comp box, in some versions of the 6137?


B. Levin
 

here's a photo of one with 'those holes' I was referring to:

http://www.ebay.com/itm/Tektronix-P6137-10X-Passive-Probe-/151225987127

my 6137 does not have that on the back.

it does not look like the user did that, as I've seen it a few times.

what is that all about?


 

That looks like a standard P6137 with the plastic cover removed.  I have several P6137 probes and they all have only one hole when fully assembled.  But if you remove the outer cover there is another base that has several adjustment holes.  I cant tell for sure but the picture you referenced appears to be a standard P6137 with the outer case removed or missing.  You can download a copy of the P6137 manual from Tektronix and check the parts diagram towards the rear.  It will show you a drawing that better describes what I think you are seeing.

As for me, I would not buy a P6137 that did not have the outer cover (one hole).  I found that when shopping on Ebay for Tektronix probes, patience is key.  I have seen P6137 probes sell for as little as $40 and as much as $200.  I paid $150 for my first pair and a week later saw one selling for $35.  

From: "snmp_4u@..."
To: TekScopes@...
Sent: Saturday, March 1, 2014 2:19 PM
Subject: [TekScopes] RE: 2400 series probes (is p6137 the most optimal?)

 
here's a photo of one with 'those holes' I was referring to:

http://www.ebay.com/itm/Tektronix-P6137-10X-Passive-Probe-/151225987127

my 6137 does not have that on the back.

it does not look like the user did that, as I've seen it a few times.

what is that all about?



r.domp.frank@...
 

That's what the compensation box looks like with its plastic cover removed. See P6137 e.a. Service Manual.


Raymond


B. Levin
 

ok, now it makes sense.  thanks guys.


lazystrings
 

For active probes you can find P6101 (will need a 1101 or 1102 power supply), P6102 (there are 2 versions, one has its own power supply, the other needs an 1101 or 1102) or you can go with the newer probes: P6205, P6243, etc. The newer probes need the 1103 power supply and is a much more expensive deal.
For low frequencies the probes shouldn't be much issue. I would have a 1X P6101. You can measure lower voltages with the 1X probe and even lower if you cascade Channel 1 and Channel 2 as explained in the manual.
I have P5050 probes that have the same tip as the P6137 but they're 500MHz probes. They work well with my TDS654C and also a 2465ADM and a 2465BDM. They have to be compensated when you use them with different scopes. 


B. Levin
 

thanks for the info on active probes.  I'm brand new to the idea of them (at least in terms of buying them).  the power supply: it looks like its just dual 15 and 5.  any reason I have to buy one as opposed to building one?  is there anything special other than the connector type?

a p6101 seems like a low end passive low freq probe.  did you really mean 6101?


B. Levin
 

what I'm mostly looking for is very high input impedance (so that I can probe things 'as if I wasn't there') and also good high freq response so that the wave that is shown is closer to what is really at the tip of the probe.

I don't want to spend $200+ on each probe.  my scope only goes to 500mhz but someday, who knows, maybe I'll end up with a 7104 or something (I was very close to buying one just a few weeks ago).  however, it needs a 50ohm probe and special plugin (vertical) to reach 1ghz.  if I ever decide to buy a scope like that, I'd like to have a probe that works with it as well as my 2400 series stuff.


 

On 02 Mar 2014 16:13:41 -0800, you wrote:

what I'm mostly looking for is very high input impedance (so that I can probe things 'as if I wasn't there') and also good high freq response so that the wave that is shown is closer to what is really at the tip of the probe.

I don't want to spend $200+ on each probe. my scope only goes to 500mhz but someday, who knows, maybe I'll end up with a 7104 or something (I was very close to buying one just a few weeks ago). however, it needs a 50ohm probe and special plugin (vertical) to reach 1ghz. if I ever decide to buy a scope like that, I'd like to have a probe that works with it as well as my 2400 series stuff.
Many active probes have surprisingly low input resistance. At high frequencies,
the shunt capacitance of a high impedance passive probe is more significant than
the its input resistance so the later tends to become unimportant. If you can
get by with more attenuation, maybe a x100 passive probe would work.

How much bandwidth do you want? Up to 100 MHz you can certainly fabricate your
own active probe. I am including an attachment showing an active probe design
from Troubleshooting Analog Circuits by Bob Pease. Page 3 of Linear Technology
application note 21 shows a similar active probe design. I would probably
combine the two:

http://www.linear.com/docs/4116

The same circuit is on page 9 of Technology application note 9. I know they
had another similar circuit somewhere using a fast FET hybrid but offhand I am
not sure where:

http://www.linear.com/docs/4105

If you do go with an active probe, they are designed to drive 50 ohm inputs so a
7104 with 7A29 will be quite at home with them assuming you can power the
probes. Most oscilloscopes faster than 100 MHz have switchable 50 ohm
terminations but for those that do not, a 50 ohm feedthrough terminatior can be
used.


John Gord
 

The P6201 and P6202A FET probes use +/15v and +5v from a LEMO connector that can plug into the probe power connector of most 7000 series scopes (not the 7603).  You can use your own supplies if you like.  The P6202 (not A) has an attached supply.  The probe power connection was an option on the 24xx scopes, consisting of a LEMO connector, a 4 wire cable, and a 4 socket connector that plugged into a 4 pin header already on the 24xx power supply board.  (Option 22 provided two such connections, iirc)

--John Gord


B. Levin
 

I did not know the 2400's had this LEMO connector as an option.  where would those mount?

if I plan to go this route, should I spend $100 on the 1101 or just DIY my own?  I'm assuming that low noise is important in this PSU.  I know of a few low noise dual 15 psu's and 5v is probably not signal-path critical so any old 5v reg would do.  do you agree?

how hard is it to get hold of those connectors that mate with the probes, for power?

would it be a good/bad idea to buy a used 1101 and swap out its whole psu and replace with something more modern?


 

You pretty much need to use a Zo (50 Ohm type) probe at high frequencies. Work out the impedance of a regular x10 probe with 10Meg in parallel with 20pF at 100MHz and you'll see that it isn't 10MegOhm, more like 80 ohms!!! At 500MHz, that's a mere 16 ohms, and by the time you get to 1GHz 8 ohms!!!

If you use a FET Probe, things get a touch better as these typically have 1pF parallel capacitance so the loading is reduced by a factor of 20.

A Zo probe is similar as the shunt capacitance is about 0.5pF.

Take a loko at the "ABCs of Probes" which is here:

<http://uk.tek.com/document/primer/abcs-probes>


Regards,
David Partridge



________________________________

From: TekScopes@yahoogroups.com [mailto:TekScopes@yahoogroups.com] On Behalf Of snmp_4u@yahoo.com
Sent: 03 March 2014 00:14
To: TekScopes@yahoogroups.com
Subject: [TekScopes] RE: 2400 series probes (is p6137 the most optimal?)




what I'm mostly looking for is very high input impedance (so that I can probe things 'as if I wasn't there') and also good high freq response so that the wave that is shown is closer to what is really at the tip of the probe.


Don Lewis <dlewis11193@...>
 

DigKey sells the mate for ~$20.00

1124-1450-ND



From: "snmp_4u@..."
To: TekScopes@...
Sent: Sunday, March 2, 2014 8:38 PM
Subject: [TekScopes] RE: 2400 series probes (is p6137 the most optimal?)

 
I did not know the 2400's had this LEMO connector as an option.  where would those mount?

if I plan to go this route, should I spend $100 on the 1101 or just DIY my own?  I'm assuming that low noise is important in this PSU.  I know of a few low noise dual 15 psu's and 5v is probably not signal-path critical so any old 5v reg would do.  do you agree?

how hard is it to get hold of those connectors that mate with the probes, for power?

would it be a good/bad idea to buy a used 1101 and swap out its whole psu and replace with something more modern?



 

On my 2440, the two Lemo probe power connectors are on the back behind the CRT
below the Lemo connector for the word recognizer.

On 02 Mar 2014 18:38:09 -0800, you wrote:

I did not know the 2400's had this LEMO connector as an option. where would those mount?

if I plan to go this route, should I spend $100 on the 1101 or just DIY my own? I'm assuming that low noise is important in this PSU. I know of a few low noise dual 15 psu's and 5v is probably not signal-path critical so any old 5v reg would do. do you agree?

how hard is it to get hold of those connectors that mate with the probes, for power?

would it be a good/bad idea to buy a used 1101 and swap out its whole psu and replace with something more modern?


B. Levin
 

$20 for that connector.  sigh.
well, I found a semi affordable power supply on ebay so took the plunge.  also found a 6202a that was affordable (and listed as 'tested') so got that too.

the psu was not much more than the raw cost of 4 connectors, so what the hell, get the tek box.  at least I know I won't swap the wrong wires (read a diy-audio post about some document being wrong in 1 out of 2 places and if you diy it, you could ruin your probe!)

so, I guess that will be my entrance into the world of active probes.  somewhat affordable, although most items listed are way overpriced, it would seem.

funny enough, I found the hook spring (witch hat) tip on amazon or the 6202a probe.  very funny.  none on the bay but one lonely one on amazon for $8 or so.  that was the main missing part from the probe that I just bought.