Re: TDS744 not 'A' has a problem , help needed

Brian
 

Hi Stephen ,
I'm pleased to hear it can all be done manually and it will be great to know how you get on with the calibration and to know your setup . When I know that I can check my gear and see if I can match the spec where its needed - for frequency accuracy if needed I have counters operating up to 40GHz should I need to use a sig gen thats not one of the synthesized ones .
Brian

On Friday, 31 May 2019, 01:32:32 BST, Stephen Hanselman <kc4sw.io@...> wrote:

Brian,

I have a 540 waiting for me after the 8566 and 8662.  Yes the process can be done manually. It’s sort of amusing that they imply automatic, but their equipment call out is all manual.

I’ll try and get my system up on Friday and get the configurations I use

Regards,

Stephen Hanselman
Datagate Systems, LLC
3107 North Deer Run Road #24
Carson City, Nevada, 89701
(775) 882-5117 office
(775) 720-6020 mobile
s.hanselman@...
www.datagatesystems.com
a Service Disabled, Veteran Owned Small Business
DISCLAIMER:
This e-mail and any attachments are intended only for use by the addressee(s) named herein and may contain legally privileged and/or proprietary information. If you are not the intended recipient, any dissemination, distribution or copying of this e-mail and any attachments is strictly prohibited. If you have received this e-mail in error, please immediately notify me and permanently delete the original and all copies and printouts of this e-mail and any attachments.
On May 30, 2019, at 14:29, Brian via Groups.Io <brianas1948=yahoo.co.uk@groups.io> wrote:

Hi Steve , how strange it needs the info in both places , I'll make notes so I dont forget that . I do have extended memory cards for the old PC's but I think I am likely to use windows 95/98 when I get around to doing this and then I'm sure it can use virtual memory on the hard disk if there is not enough main memory - I still have copies of s/w like ramdoubler if things get desperate .I have copies of the cal s/w from a guy on EEblog who put up the files for 500,600,700 series scopes , I have not run them yet so I hope there are no problems .Can the whole process be done manually as I am more likely to try that because the instruments I have are almost certainly not compatible with the GPIB cal s/w , some do not have a GPIB port at all .
Its good to speak to someone who has done it as it makes it feel so much more possible for me to do it
Brian

    On Thursday, 30 May 2019, 21:58:02 BST, Stephen Hanselman <kc4sw.io@...> wrote:

I suspect you have a better shot at it than I did.  One other thing when setting up the PC set files and buffers up a good bit also add some more memory above the base 640K.  also the config file for the cal software has a line that alleges to define the place it looks for the HPIB card.  It has two possible "places", I'll look later, and you need to put the card info into both of them.  This even happened on my Tek DM system.


Have you got the cal program files?

steve

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Brian via Groups.Io
Sent: Thursday, May 30, 2019 1:06 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] TDS744 not 'A' has a problem , help needed

Hi Steve , thank you for that . I have heard several ideas about the NVRAM contents , one is said to hold just saved waveforms and setups and the other holding cal data and installed options , cant remember which is said to do what so I'll need to search EEblog again . I have heard it said also that there are a couple of NVRAM chips underneath the acquisition pcb , little 8 pin chips , that hold factory cal data but that it is only a copy of one of the Dallas chips and is there for emergency's for use at a cal center , those 2 chips are there even on my non A version . The same source if I remember correctly said that the system does not actually use this data it relies on the Dallas chip .
I have a Fluke DC calibrator and 6.5,7.5 and 8.5 digit DVM's. Signal generators , I have synthesized gen's up to 18GHz but their output is too low , but I do have a couple of small  1MHz - 5GHz amps with output up to 20dBm that might make it possible . the final output will not be levelled directly but the input will be , to make certain I do have HP438A power meter and sensors , I also have a RACAL true RMS voltmeter that is good to 2GHz .Unlike some here I do have several old PC's with windows 95/98 could even run windows 3.11 , I have also the NI PC2A GPIB cards and drivers and a copy of the cal s/w

I am in the UK if you hadnt guessed so I think we are more than just a few miles apart or I would be happy to join you for that after school project Brian

    On Thursday, 30 May 2019, 20:32:31 BST, Stephen Hanselman <kc4sw.io@...> wrote:

I've heard the talk flow both ways but to my knowledge the calibration data is held in NVRAM on the Acquisition pca not in the Proc.  I've looked in several of the Dallas Chips and mostly what you see is empty space.  I suspect, but have no proof, that these things are used to store trace data.  I have taken non-working (fails SPC etc) and swapped in working calibrated Acq and had it remember it's calibration data.

Calibration requirements are actually fairly lax.  You need a good, accurate DC source, DMM, and after that almost any reasonable sig gen would work.  I have SG503 and SG504 the scope tells you if the particular signal you are looking at from the "SG's" is correct and then goes off and calibrates itself.  The initial DC calibrations are critical  however.  The las thing is a HPIB based controller to run the test suite.  I tried several incarnations and finally bought a Tek Data Manager.  Given no experience and all manual TE I would plan on a full day for the cal, and be prepared to become extremely bored.

If you look at your diagnostics screen where it shows SPC, Voltage ref, Frequency...... there are 4 states Initialized (after cold start), running, passed, failing.  I found out the hard way that even when you pass the CVR testing voltage ref may say "running"  although not next in the sequence you have to run probe comp to get it to read "passing"  there are a few other gotchas in the software similar to this.

If you were close by I'd tell you to bring it in as an afterschool project, but...

Hope this helps you a bit.  Like they say though, your mileage may vary

steve

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Brian via Groups.Io
Sent: Thursday, May 30, 2019 8:19 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] TDS744 not 'A' has a problem , help needed

Hi Stephen , fortunately the scope remembers the date and the clock is running fine . I have NVRAM back-up s/w from EEBLOG so hopefully it will keep a record of the contents assuming it works on the much older scopes but I cant see why it shouldnt . I do have some GPIB experience and enough gear to connect to the scope .The scope does have nice options for me 13,1f,2f,2m so I dont really want to loose any of themThe Dallas chips are DS1650Y and DS1486 , I understand that one of them holds calibration data - I dont know that even with the gear I have a complete recalibration would be possible so I need to keep that info as intact as possible for as long as I can . Any calibration equipment that I dont currently have will take time to get hold of as I am retired and only a hobbyist/collector .

    On Thursday, 30 May 2019, 16:03:41 BST, Stephen Hanselman <kc4sw.io@...> wrote:

Brian,

Changing out these chips is not really a problem. While they do keep some scope specific info I’ve operated 540A’s without the chip.  If you have a vanilla model, no options I’m not sure it’s worth the effort.  EBay, yes I know, has some replacements, when you look at the offerings look for one that has a “+” in the part number. So it seems like these are newer ones. The Tek forum has a thread that gives the cross wiring needed to program the Dallas chip in a more common NVRAM environment.

On the other hand if your scope remembers the date and time the chip is probably ok, for a while.

My technique is to add fresh solder to each pin and then heat the pin, not the board, and suck it out.  I then insert a socket.  So far I have not had an issue with losing data in the chip.

As it stands I have good images for the 540, 540A, and 540B.

Regards,

Stephen Hanselman
Datagate Systems, LLC
3107 North Deer Run Road #24
Carson City, Nevada, 89701
(775) 882-5117 office
(775) 720-6020 mobile
s.hanselman@...
www.datagatesystems.com
a Service Disabled, Veteran Owned Small Business
DISCLAIMER:
This e-mail and any attachments are intended only for use by the addressee(s) named herein and may contain legally privileged and/or proprietary information. If you are not the intended recipient, any dissemination, distribution or copying of this e-mail and any attachments is strictly prohibited. If you have received this e-mail in error, please immediately notify me and permanently delete the original and all copies and printouts of this e-mail and any attachments.
On May 30, 2019, at 07:53, Brian via Groups.Io <brianas1948=yahoo.co.uk@groups.io> wrote:

Hi Bob , thank you . That was only the second one of those scopes I've had anything to do with , I am mostly 500 series + 661 and 564 along with 7000 series and 24XX .  My first one was a TDS420A with a dead PSU , changing the caps revived it , nothing else was wrong with it . The TDS744 was also one that did not start up , after sorting out the PSU the other faults showed themselves , it was the first time fixing a component level fault on an acquisition pcb . I now have to do something about the Dallas chips they are dated '93!!

    On Thursday, 30 May 2019, 15:26:46 BST, Bob Koller via Groups.Io <testtech=yahoo.com@groups.io> wrote:

Well done!



















Join TekScopes@groups.io to automatically receive all group messages.