Topics

2454B 400MHz gives just about 300 MHz BW


Ulli Wiemers
 

Hi,

my 2465B gives just 1.09 nsec fall time and 1.23 nsec rise time with a directly connected <Fast risetime (40ps) pulse generator> (http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=124&products_id=295).
CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(

Any idea what could make that device so slow?

Thanks
Ulli


cmjones01
 

On 30 Sep 2017 12:57 p.m., "modmix modmix@... [TekScopes]" <
TekScopes@...> wrote:



Hi,

my 2465B gives just 1.09 nsec fall time and 1.23 nsec rise time with a
directly connected <Fast risetime (40ps) pulse generator>
(http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=124&
products_id=295).
CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(

Any idea what could make that device so slow?

Do you have the scope input set to 50R? If it's on the normal high
impedance setting I would expect it to be a bit slower. I would only expect
any fast scope input only to give its best performance with the correct
termination. Even with the pulser plugged directly into the scope's input,
the few centimetres of distance between the source and scope make a
difference at these sort of speeds if there are any reflections.

Chris


Ulli Wiemers
 

Yes, not using 50R at the scope's input is a nice trap - of course I did run directly into it at the first attempt ,-)
From that I know: 1.47 rise time (~ 240 MHz) - too little to be true...
=> measurement was done using 50R termination
Ulli


Bob Albert
 

Are you terminating the line at the oscilloscope?  Otherwise you may have standing waves on the cable, resulting in resonances, etc.
Bob

On Saturday, September 30, 2017 6:34 AM, "modmix modmix@... [TekScopes]" <TekScopes@...> wrote:


  Yes, not using 50R at the scope's input is a nice trap - of course I did
run directly into it at the first attempt ,-)
From that I know: 1.47 rise time (~ 240 MHz) - too little to be true...
=> measurement was done using 50R termination
Ulli
#yiv9636770477 #yiv9636770477 -- #yiv9636770477ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv9636770477 #yiv9636770477ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv9636770477 #yiv9636770477ygrp-mkp #yiv9636770477hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv9636770477 #yiv9636770477ygrp-mkp #yiv9636770477ads {margin-bottom:10px;}#yiv9636770477 #yiv9636770477ygrp-mkp .yiv9636770477ad {padding:0 0;}#yiv9636770477 #yiv9636770477ygrp-mkp .yiv9636770477ad p {margin:0;}#yiv9636770477 #yiv9636770477ygrp-mkp .yiv9636770477ad a {color:#0000ff;text-decoration:none;}#yiv9636770477 #yiv9636770477ygrp-sponsor #yiv9636770477ygrp-lc {font-family:Arial;}#yiv9636770477 #yiv9636770477ygrp-sponsor #yiv9636770477ygrp-lc #yiv9636770477hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv9636770477 #yiv9636770477ygrp-sponsor #yiv9636770477ygrp-lc .yiv9636770477ad {margin-bottom:10px;padding:0 0;}#yiv9636770477 #yiv9636770477actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv9636770477 #yiv9636770477activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv9636770477 #yiv9636770477activity span {font-weight:700;}#yiv9636770477 #yiv9636770477activity span:first-child {text-transform:uppercase;}#yiv9636770477 #yiv9636770477activity span a {color:#5085b6;text-decoration:none;}#yiv9636770477 #yiv9636770477activity span span {color:#ff7900;}#yiv9636770477 #yiv9636770477activity span .yiv9636770477underline {text-decoration:underline;}#yiv9636770477 .yiv9636770477attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv9636770477 .yiv9636770477attach div a {text-decoration:none;}#yiv9636770477 .yiv9636770477attach img {border:none;padding-right:5px;}#yiv9636770477 .yiv9636770477attach label {display:block;margin-bottom:5px;}#yiv9636770477 .yiv9636770477attach label a {text-decoration:none;}#yiv9636770477 blockquote {margin:0 0 0 4px;}#yiv9636770477 .yiv9636770477bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv9636770477 .yiv9636770477bold a {text-decoration:none;}#yiv9636770477 dd.yiv9636770477last p a {font-family:Verdana;font-weight:700;}#yiv9636770477 dd.yiv9636770477last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv9636770477 dd.yiv9636770477last p span.yiv9636770477yshortcuts {margin-right:0;}#yiv9636770477 div.yiv9636770477attach-table div div a {text-decoration:none;}#yiv9636770477 div.yiv9636770477attach-table {width:400px;}#yiv9636770477 div.yiv9636770477file-title a, #yiv9636770477 div.yiv9636770477file-title a:active, #yiv9636770477 div.yiv9636770477file-title a:hover, #yiv9636770477 div.yiv9636770477file-title a:visited {text-decoration:none;}#yiv9636770477 div.yiv9636770477photo-title a, #yiv9636770477 div.yiv9636770477photo-title a:active, #yiv9636770477 div.yiv9636770477photo-title a:hover, #yiv9636770477 div.yiv9636770477photo-title a:visited {text-decoration:none;}#yiv9636770477 div#yiv9636770477ygrp-mlmsg #yiv9636770477ygrp-msg p a span.yiv9636770477yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv9636770477 .yiv9636770477green {color:#628c2a;}#yiv9636770477 .yiv9636770477MsoNormal {margin:0 0 0 0;}#yiv9636770477 o {font-size:0;}#yiv9636770477 #yiv9636770477photos div {float:left;width:72px;}#yiv9636770477 #yiv9636770477photos div div {border:1px solid #666666;min-height:62px;overflow:hidden;width:62px;}#yiv9636770477 #yiv9636770477photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv9636770477 #yiv9636770477reco-category {font-size:77%;}#yiv9636770477 #yiv9636770477reco-desc {font-size:77%;}#yiv9636770477 .yiv9636770477replbq {margin:4px;}#yiv9636770477 #yiv9636770477ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv9636770477 #yiv9636770477ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv9636770477 #yiv9636770477ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv9636770477 #yiv9636770477ygrp-mlmsg select, #yiv9636770477 input, #yiv9636770477 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv9636770477 #yiv9636770477ygrp-mlmsg pre, #yiv9636770477 code {font:115% monospace;}#yiv9636770477 #yiv9636770477ygrp-mlmsg * {line-height:1.22em;}#yiv9636770477 #yiv9636770477ygrp-mlmsg #yiv9636770477logo {padding-bottom:10px;}#yiv9636770477 #yiv9636770477ygrp-msg p a {font-family:Verdana;}#yiv9636770477 #yiv9636770477ygrp-msg p#yiv9636770477attach-count span {color:#1E66AE;font-weight:700;}#yiv9636770477 #yiv9636770477ygrp-reco #yiv9636770477reco-head {color:#ff7900;font-weight:700;}#yiv9636770477 #yiv9636770477ygrp-reco {margin-bottom:20px;padding:0px;}#yiv9636770477 #yiv9636770477ygrp-sponsor #yiv9636770477ov li a {font-size:130%;text-decoration:none;}#yiv9636770477 #yiv9636770477ygrp-sponsor #yiv9636770477ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv9636770477 #yiv9636770477ygrp-sponsor #yiv9636770477ov ul {margin:0;padding:0 0 0 8px;}#yiv9636770477 #yiv9636770477ygrp-text {font-family:Georgia;}#yiv9636770477 #yiv9636770477ygrp-text p {margin:0 0 1em 0;}#yiv9636770477 #yiv9636770477ygrp-text tt {font-size:120%;}#yiv9636770477 #yiv9636770477ygrp-vital ul li:last-child {border-right:none !important;}#yiv9636770477



[Non-text portions of this message have been removed]


cmjones01
 

On 30 Sep 2017 3:34 p.m., "modmix modmix@... [TekScopes]" <
TekScopes@...> wrote:



Yes, not using 50R at the scope's input is a nice trap - of course I did
run directly into it at the first attempt ,-)
From that I know: 1.47 rise time (~ 240 MHz) - too little to be true...
=> measurement was done using 50R termination

Interesting. Glad you spotted that one!

Then I can't think of a good explanation for why your scope appears to be
slow. I checked my 2465A and it came up to spec, with a 1ns risetime.

Here are a few thoughts which might be clutching at straws:
- try different attenuator settings
- I hope I'm not teaching my grandmother to suck eggs, but you are
measuring from the 10% to 90% amplitude points, aren't you?
- there was some story on here a few years ago about there being some 2465s
around which were unofficially "upgraded" 2445s. Maybe you've got one of
those? It's a very remote possibility though.

Chris


Ulli Wiemers
 

There is no cable or so involved. The device is attached directly to the bnc input set to 50R.
The link in my first posting in this thread has some pictures showing how the device is attached.
Ulli


TomC
 

For comparison, I get 0.89 nsec rise time with my 2465 using the same pulse generator. Your 2465B should be better than my 2465.

Are you measuring the time between 10% and 90%?

Tom

On 9/30/2017 3:54 AM, modmix modmix@... [TekScopes] wrote:
Hi,
my 2465B gives just 1.09 nsec fall time and 1.23 nsec rise time with a
directly connected <Fast risetime (40ps) pulse generator>
(http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=124&products_id=295).
CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(
Any idea what could make that device so slow?
Thanks
Ulli


Sergey Kubushyn
 

On Sat, 30 Sep 2017, Tom tomc@... [TekScopes] wrote:

Just a simple stupid question -- is that 2465B calibrated?

BTW, _THAT_ part of calibration is trickiest of them all.

For comparison, I get 0.89 nsec rise time with my 2465 using the same pulse
generator. Your 2465B should be better than my 2465.

Are you measuring the time between 10% and 90%?

Tom




On 9/30/2017 3:54 AM, modmix modmix@... [TekScopes] wrote:
Hi,

my 2465B gives just 1.09 nsec fall time and 1.23 nsec rise time with a
directly connected <Fast risetime (40ps) pulse generator>
(http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=124&products_id=295).
CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(

Any idea what could make that device so slow?

Thanks
Ulli
---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Ulli Wiemers
 

Are you measuring the time between 10% and 90%?
Yes.

In the EEVblog thread
Re: Yet another fast edge pulse generator
(http://www.eevblog.com/forum/projects/yet-another-fast-edge-pulse-generator/msg1313591/#msg1313591)
I've attached a screens shot showing the automatic measurement result:
http://www.eevblog.com/forum/projects/yet-another-fast-edge-pulse-generator/?action=dlattach;attach=355994;image.

Using grid scale (0-10-90-100%) gives more or less the same figures.

Ulli

BTW:
Any chance to post images here?


Ulli Wiemers
 

"calibrated?" - good question.
I got this device a couple of years ago - will try to ask seller and report back.

FYI:
Automatic frequency measurement gives 10.24 MHz.
GPS controlled Agilent 53230A gives 10.000... MHz.
=> scope measures frequency better the 3%.
Bandwidth is about 25% too low.
Are these to topics related to each other?

Ulli


Sergey Kubushyn
 

On Sat, 30 Sep 2017, modmix modmix@... [TekScopes] wrote:

"calibrated?" - good question.
I got this device a couple of years ago - will try to ask seller and
report back.
You will have to go through part 6 (if my dementia serves me right -- it's
been couple of years since I calibrated 24xx scope last time) of calibration
routine. It is bandwidth calibration routine that is exactly for achieving
full bandwidth.

FYI:
Automatic frequency measurement gives 10.24 MHz.
GPS controlled Agilent 53230A gives 10.000... MHz.
=> scope measures frequency better the 3%.
Bandwidth is about 25% too low.
Are these to topics related to each other?
Nope, they have nothing in common.

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


Chuck Harris
 

Lot's of things could make your scope test slow, but frequently
it is just calibration.

There are some fake 2465B's out there that are actually
2445B's that have had their frequency limiting, spiral delay
line matching, network cut out, and the mother board jumpered
so that the CPU (A5) controller thinks it is working with a
2465B. They also generally have a shiny aluminum stick-on
label over the CRT, with blue lettering. Many of these were
done and sold by a person named Alex S., whose ebay ID is
2465b. If you bought a 2465b scope from Alex S., it is quite
probably a fake.

I have one in my collection that Alex S. acknowledged
creating. He admitted the whole scheme, and was quite
unrepentant about his fraud. My scope came out of a US
government DOD scrap pile. It never could meet the risetime
spec that was required of a real 2465B, but was close to
meeting that required by the older 300MHz 2465.

As far as I can tell, Alex S. did hundreds, if not thousands.

-Chuck Harris


modmix modmix@... [TekScopes] wrote:

CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(

Any idea what could make that device so slow?

Thanks
Ulli


------------------------------------
Posted by: modmix <modmix@...>
------------------------------------


Dave Daniel
 

Thanks, Chuck.

I looked at this guy's current listings. All his prices seem a little
high, and many of them state something like "will sell for more of you
want it to be calibrated". Most of the listings turned my "eBay BS
radar" on.

BTW, my 2465B that you calibrated is still my go-to 'scope.

DaveD

On 9/30/2017 2:43 PM, Chuck Harris cfharris@... [TekScopes] wrote:

Lot's of things could make your scope test slow, but frequently
it is just calibration.

There are some fake 2465B's out there that are actually
2445B's that have had their frequency limiting, spiral delay
line matching, network cut out, and the mother board jumpered
so that the CPU (A5) controller thinks it is working with a
2465B. They also generally have a shiny aluminum stick-on
label over the CRT, with blue lettering. Many of these were
done and sold by a person named Alex S., whose ebay ID is
2465b. If you bought a 2465b scope from Alex S., it is quite
probably a fake.

I have one in my collection that Alex S. acknowledged
creating. He admitted the whole scheme, and was quite
unrepentant about his fraud. My scope came out of a US
government DOD scrap pile. It never could meet the risetime
spec that was required of a real 2465B, but was close to
meeting that required by the older 300MHz 2465.

As far as I can tell, Alex S. did hundreds, if not thousands.

-Chuck Harris

modmix modmix@... [TekScopes] wrote:

CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(

Any idea what could make that device so slow?

Thanks
Ulli


------------------------------------
Posted by: modmix <modmix@...>
------------------------------------


johnhstrauch
 

Yea, I got somewhat screwed by this guy a couple of years ago through ebay. 2465b clearance sale!!!! It lasted a couple of months of minimal usage and by that I mean maybe turned on for 12 hours total over that time. Then the A5 board failed with the all too familiar leaky cap damage issues. I opened it up and the caps had been changed to tantalums but I could not really check their pad / solder condition due to the very hard potting placed around them and their traces. When I contacted the quack Alex he said I could send it in with $600 for repair. From his emails I deduced he is an arse. He did answer my couple of subsequent emails with a bit of tech info though.

Long story short I bought a newer A5 board and replaced and calibrated and it works good.


Is there a way to identify the A1 board that is 2445B vs. 2465b without the considerable hassle of removal to do a visual !


When that loose screw fell out from under the A5 board I had that gut feeling. Bad gut feeling.


I actually have not done any bandwidth checks.


Just my story.


JH


Ulli Wiemers
 

SK> You will have to go through part ... of calibration routine.
SK> It is bandwidth calibration routine that is exactly for achievingfull bandwidth.
SK> Nope, they have nothing in common.
Glad to be here - I learn a lot. Thanks!

CH> Lot's of things could make your scope test slow, but frequently it is just calibration.
Sound like good news

Seller responded: device was calibrated when I got it about 5 years ago.
Cloud 5 years be long long enough to degrade bandwidth by 25%?...

CH> There are some fake 2465B's out there that are actually 2445B's that have had their frequency limiting,...
Would the serail number (B024640) help to tell the version?

CH> sold by a person named Alex S.
That's not the person I got it from.
But admitted, I don't know its entire history...

Seller offered to check the device. I'm invited to join. Will most likely happen end of this / beginning of next year.
I'll report back the findings.

Thanks so far for all your input.
Ulli


Ulli Wiemers
 

JH> Is there a way to identify the A1 board that is 2445B vs. 2465b
My 2465B has A1 PCB part number G9A-0037-01 671-0722-08.

tek-forum - Is a modded 2445B a 2465B? https://forum.tek.com/viewtopic.php?t=136149> to a real 671-0722-08 2465B A1 PCB
=> looks to be a real 2465B

Ulli


Sergey Kubushyn
 

On Sat, 30 Sep 2017, johnhstrauch@... [TekScopes] wrote:

Yea, I got somewhat screwed by this guy a couple of years ago through
ebay. 2465b clearance sale!!!! It lasted a couple of months of minimal
usage and by that I mean maybe turned on for 12 hours total over that
time. Then the A5 board failed with the all too familiar leaky cap damage
issues. I opened it up and the caps had been changed to tantalums but I
could not really check their pad / solder condition due to the very hard
potting placed around them and their traces. When I contacted the quack
Alex he said I could send it in with $600 for repair. From his emails I
deduced he is an arse. He did answer my couple of subsequent emails with
a bit of tech info though.

Long story short I bought a newer A5 board and replaced and calibrated and
it works good.


Is there a way to identify the A1 board that is 2445B vs. 2465b without
the considerable hassle of removal to do a visual !
It is easy. Real 2465B A1 has 3 potentiometers next to U900 that are
"HrzCtr", "X10 Gain", and "Transient Response". Fake one doesn't have
"Transient Response" pot; there are only two there. And guess what -- that
"Transient Response" potentiometer is a part of calibration routine for
getting 400MHz bandwidth! :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************


bobh@joba.com
 

I generally skip past ebay offerings with generic pictures rather than the actual instrument. And I want to see a serial number so I can verify what I receive is what was in the pictures. The seller mentioned fails on both of these tests.
Bob.

From: Dave Daniel kc0wjn@... [TekScopes]
Sent: Saturday, September 30, 2017 1:59 PM
To: TekScopes@...
Subject: Re: [TekScopes] 2454B 400MHz gives just about 300 MHz BW


Thanks, Chuck.

I looked at this guy's current listings. All his prices seem a little
high, and many of them state something like "will sell for more of you
want it to be calibrated". Most of the listings turned my "eBay BS
radar" on.

BTW, my 2465B that you calibrated is still my go-to 'scope.

DaveD

On 9/30/2017 2:43 PM, Chuck Harris cfharris@... [TekScopes] wrote:

Lot's of things could make your scope test slow, but frequently
it is just calibration.

There are some fake 2465B's out there that are actually
2445B's that have had their frequency limiting, spiral delay
line matching, network cut out, and the mother board jumpered
so that the CPU (A5) controller thinks it is working with a
2465B. They also generally have a shiny aluminum stick-on
label over the CRT, with blue lettering. Many of these were
done and sold by a person named Alex S., whose ebay ID is
2465b. If you bought a 2465b scope from Alex S., it is quite
probably a fake.

I have one in my collection that Alex S. acknowledged
creating. He admitted the whole scheme, and was quite
unrepentant about his fraud. My scope came out of a US
government DOD scrap pile. It never could meet the risetime
spec that was required of a real 2465B, but was close to
meeting that required by the older 300MHz 2465.

As far as I can tell, Alex S. did hundreds, if not thousands.

-Chuck Harris

modmix modmix@... [TekScopes] wrote:

CH1 & CH2 are almost identical.
That's about just 300 MHz... ,-(

Any idea what could make that device so slow?

Thanks
Ulli


------------------------------------
Posted by: modmix <modmix@...>
------------------------------------


Chuck Harris
 

A simple visual will tell the tale. The 2445B has two spiral inductors that
are towards the rear of the scope (1/4 from the back), in the middle (eg
left/right). There is a "G" drawn on the board in the vicinity, and it is
also near where the delay line peeks up through the board and makes its
connection. You can see the spiral inductors as they show through the layers
and solder mask.

There are two additional pots on the original, and another coil, but the
spiral inductors are the clearest indication.

I have had customers ask me to try to make their fake scope better. I have
added the two missing pots, and to do an extra special job on adjusting the
transient response, but the missing delay line matching network makes it
impossible to get all the way to a 2465B in performance.

-Chuck Harris

modmix@... [TekScopes] wrote:

JH> Is there a way to identify the A1 board that is 2445B vs. 2465b My 2465B has
A1 PCB part number G9A-0037-01 671-0722-08.

tek-forum - Is a modded 2445B a 2465B?
https://forum.tek.com/viewtopic.php?t=136149> to a real 671-0722-08 2465B A1 PCB
=> looks to be a real 2465B

Ulli




Chuck Harris
 

Look harder, there is another pot, and a missing inductor.

-Chuck Harris

Sergey Kubushyn @KSI [TekScopes] wrote:

On Sat, 30 Sep 2017, johnhstrauch@... [TekScopes] wrote:

Yea, I got somewhat screwed by this guy a couple of years ago through
ebay. 2465b clearance sale!!!! It lasted a couple of months of minimal
usage and by that I mean maybe turned on for 12 hours total over that
time. Then the A5 board failed with the all too familiar leaky cap damage
issues. I opened it up and the caps had been changed to tantalums but I
could not really check their pad / solder condition due to the very hard
potting placed around them and their traces. When I contacted the quack
Alex he said I could send it in with $600 for repair. From his emails I
deduced he is an arse. He did answer my couple of subsequent emails with
a bit of tech info though.

Long story short I bought a newer A5 board and replaced and calibrated and
it works good.


Is there a way to identify the A1 board that is 2445B vs. 2465b without
the considerable hassle of removal to do a visual !
It is easy. Real 2465B A1 has 3 potentiometers next to U900 that are
"HrzCtr", "X10 Gain", and "Transient Response". Fake one doesn't have
"Transient Response" pot; there are only two there. And guess what -- that
"Transient Response" potentiometer is a part of calibration routine for
getting 400MHz bandwidth! :)

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************