Date   

Re: Tek 4041 GPIB Controller

Monty McGraw
 

I have spent the last couple of weeks with the EZ-TEST Generator (Volume 1) and Translator (Volume 2) programs and can confirm the files I posted are complete and work fine.

The EZ-TEST Generator program does require a Tektronix 4105 / 4107 color graphics terminal - but I found a VersaTerm Pro emulator program for the 68K Mac that runs in a Mac emulator called Basilisk II on Windows. VersaTerm Pro emulates DEC VT and Tektronix graphics terminals including the 4105.

EZ-TEST Generator programs must be loaded and run with the Generator, but the Translator can create a standalone 4041 program from an EZ-TEST program that runs with a regular serial console and is faster than the EZ-TEST scripts.

My 4041 forum photo album has photos of EZ-TEST running on the VersaTerm Pro emulator: https://groups.io/g/TekScopes/photo/247590/0?p=Created,,,20,2,0,0

I created a simple two-step EZ-TEST program with the Generator, then used the Translator to convert it to a BASIC program. I was then able to edit that BASIC program to add commands that I didn't find in Generator menus.

I just posted info on vcfed.org http://www.vcfed.org/forum/showthread.php?78257-Tektronix-4041-EZ-Test-tapes-recovered!!&p=652071#post652071 on my first EZ-TEST program, which runs some tests on my Tektronix MI5010 and six interface cards: ADC (analog to digital converter), DAC (digital to analog converter), DIO (digital I/O), RSC (relay switches) and a MEMORY card that can DMA digital data from the other cards into buffer memory, faster than GPIB transfers.


Re: Fabricating "blank" TM500 Modules

Harvey White
 

If you want GPIB control, then you need TM5000.

If you want some of the functions that aren't available in the TM500 series, then you need TM5000.

If you don't need either, the go for TM500.

Perhaps a mixture of both?

Harvey

On 1/21/2021 7:19 PM, radeng wrote:
There is usually no swap meet/eBay premium for TM5003 and 5006 mainframes over the TM500 equivalents. They are an additional 1RU high, so if rack space is a consideration, go with the TM506. GPIB is nice, but the old 4041 controllers use a tape drive and no one seems to have a good tape.





Re: NPN-PNP diff pairs; Was 2235 Horizontal Calibration Issue

Dave Peterson
 

Ozan,
I'm an HSpice jockey. I'd love a look at your LTSpice model and come up to speed on component simulation (vs. IC MOS simulation).
Dave

On Thursday, January 21, 2021, 04:05:35 PM PST, Ozan <ozan_g@...> wrote:

I put together a representative LTSpice schematic. Sim shows the optimized ramp-up response, most of the current drive at highest ramp (5ns/div with x10 mag) is supplied by the caps at the emitters. It is also interesting that the NPN/PNP pair is good old 2n3904/2n3906, the circuit gets the last drop of performance out of them.

If there is interest I can post the LTspice schematic.
Ozan


Re: Fabricating "blank" TM500 Modules

 

There is usually no swap meet/eBay premium for TM5003 and 5006 mainframes over the TM500 equivalents. They are an additional 1RU high, so if rack space is a consideration, go with the TM506. GPIB is nice, but the old 4041 controllers use a tape drive and no one seems to have a good tape.


Re: TEK465 Horizontal trace

Mlynch001
 

I would also check Q1274, Q1282 and Q1288 as well as all the diodes on that left side of the circuit. One good thing is that you have one side of the circuit working. You can use the “good” side as a reference for what good parts should test like.

--
Michael Lynch
Dardanelle, AR


Re: NPN-PNP diff pairs; Was 2235 Horizontal Calibration Issue

Ozan
 

I put together a representative LTSpice schematic. Sim shows the optimized ramp-up response, most of the current drive at highest ramp (5ns/div with x10 mag) is supplied by the caps at the emitters. It is also interesting that the NPN/PNP pair is good old 2n3904/2n3906, the circuit gets the last drop of performance out of them.

If there is interest I can post the LTspice schematic.
Ozan


Re: TEK465 Horizontal trace

Ozan
 

Hi Wes,
We were looking at horizontal output stages for another thread so this one is another good schematic to look at. As others also commented Q1234 and Q1224 function as single ended to differential converter and single ended signal is applied to Q1224. We expect the same signal at the emitters (they are connected after all), base of Q1224 will have signal but base of Q1234 will not have any signal (or very tiny). Your observation of no signal at B of Q1234 is expected. Your reading of 24 ohms from base to ground is also reasonable (should be lower but I don't think this is the problem).

Not seeing a signal similar (but opposite polarity) at collector of Q1234 while you are seeing a signal at collector of Q1224 is a problem. You didn't explicitly say you are seeing good signal at collector of Q1224, do you see a good signal?

The other thing I found out is the first service manual I pulled from W140 didn't have voltages. This one has:
https://w140.com/tekwiki/images/c/ca/070-1861-00_C.pdf

Could you report all the DC voltages at E,B,C of Q1234 and Q1224?

As Michael wrote I heard 465 has sockets for transistors. You may want to wiggle and re-seat the transistors (with power off) for all the transistors on page 222 and 223 of the link I gave.
Ozan


Re: TEK465 Horizontal trace

Michael W. Lynch
 

Wes,

I don't suppose you have a working scope? I know that you already know this, but this is where having a scope shines, nothing beats looking for the correct wave forms in a circuit.

Are your Transistors soldered or are they in sockets? Most 465's are in sockets. I have a similar issue and was able to swap transistors from one side of this circuit and find the problem. There are several interchangeable transistors in that circuit. You can swap these components and see if the problem "jumps" from one side to the other.

I also had a 465 that would not trigger and it turned out that I had a very weak transistor (Q552 on the B Trigger Gen) in the circuit. I found this problem by swapping transistors from the working "A" Trigger Circuit.

Of course, you need to proceed with caution and make 100% certain that you are "Swapping" the right parts. Your Replaceable Parts list in the Service Manual will help you identify the parts that are interchangeable,

Good Luck.

--
Michael Lynch
Dardanelle, AR


Re: 2235 Horizontal Calibration Issue

Stephen
 

On Thu, Jan 21, 2021 at 11:32 AM, fiftythreebuick wrote:


Sorry for the bother....
Absolutely no bother at all, on the contrary.
Thank you for your input....


Re: 2235 Horizontal Calibration Issue

fiftythreebuick
 

Sorry for the bother....


Re: 2235 Horizontal Calibration Issue

Stephen
 

On Thu, Jan 21, 2021 at 10:43 AM, fiftythreebuick wrote:


Hi Stephen-

Have you checked for the horizontal non-linearity with the scale illumination
switched off? Just a thought....

Good luck!

Tom
Of course I have. I’ve just installed it a few days ago. This scope has always had that issue since I have it.


Re: TDS5054 infos...

mill_glen_chevy
 

Hi gents'

I'd like to add the options to mine as well. Would someone please pass along the info to add them?

Thanks

--
Brian St Pierre


Re: 3 Books still sitting in USPS NETWORK DISTRIBUTION CENTERS

Michael A. Terrell
 

Good luck with Fedex. They keep leaving my stuff at the wrong address. One
package was lost for over six months. The irony is that that package was
new hard drives for my video surveillance system.

On Thu, Jan 21, 2021 at 10:45 AM kim.herron@... <
kim.herron@...> wrote:

I work for USPS. The machines that were taken out of
service, in many cases, are machines that were slated to
be retired 5 to 8 years ago. They've been setting in plants
unused and slated for disposal for a while. There is one
whole plant in Kalamazoo that's been setting unused for
at least 5 years.
The short story is that between the virus issues and
upper management not permitting the hiring of temp
workers and paying extra for full time employees to get
stuff done, they can't get stuff moved. There are
THOUSANDS of semi-trailers all over the country with
mail setting in them
that isn't being sorted and moved. It varies around the
country. It's not the machinery that's the issue. It's that
there isn't enough people and money to get the job done.
IF you are sending ANYTHING through USPS, it needs to
be Priority, Express, or registered. Anything else is a
crap shoot. I've been using FEDEX to things shipped
because they at least are hiring people and their costs
are not as high as UPS.



On 21 Jan 2021 at 5:12, Michael A. Terrell wrote:

Weren't the machines that they retired from the first generation,
that were
installed 30 years ago? I remember them being installed in Central
Florida
at that time. We got frantic phone calls from the sorting center
looking
for some very specific trim pots to repair the brand new machines
before
they could be pressed into service. That generation of machines was
always
problematic, and should have been replaced years ago. I was working
forr a
resource recovery business who wholesaled surplus parts and
equipment to
the area's surplus dealers. Some of the components were already EOL,
when
the machines were built.

On Tue, Jan 19, 2021 at 7:29 PM <WCOFFIN1@...> wrote:

the postal system has been trashed to some degree, many of their
high speed
sorters were taken out of service last year, I didn't get my
utility bill
here until after it was due this time.
The usps uses a priority system when loading shipments within
their system
and "media mail" are lowest priority (and the lowest cost) so they
will sit
and wait until most of the others of are shipped out.

<
https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm
_campaign=sig-email&utm_content=webmail&utm_term=icon
Virus-free.
www.avast.com
<
https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm
_campaign=sig-email&utm_content=webmail&utm_term=link
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Tue, Jan 19, 2021 at 12:20 AM Dennis Tillman W7pF
<dennis@...>
wrote:

This book has been in the PHILADELPHIA PA NETWORK DISTRIBUTION
CENTER
since
Jan 8 and it hasn't moved.
Tracking #: 9549016335941002722853, Bryn Mawr PA 19010

These two books are still in the SEATTLE WA NETWORK DISTRIBUTION
CENTER
since I mailed them on Jan 2.
Tracking #: 9549016335941002723034, Wichita, KS 67212
Tracking #: 9549016335941002723140, Reno, NV 89511

Dennis Tillman W7pF












Kim Herron W8ZV
kim.herron@...
1-616-677-3706







Re: 2235 Repair having no power

Roy Thistle
 

On Thu, Jan 21, 2021 at 08:20 AM, satbeginner wrote:


After taking it out, I saw the little glass particles inside the CRT.
Did you see... or do you know what they are from. ( I'm always curious about how things fail... and about CRTs in general.) Regards.


Re: 2235 Horizontal Calibration Issue

fiftythreebuick
 

Hi Stephen-

Have you checked for the horizontal non-linearity with the scale illumination switched off? Just a thought....

Good luck!

Tom


Re: 2465b NVRAM replacement -2020?

Chuck Harris <cfharris@...>
 

Hi Mark,

The 2445B uses the same CPU card as the 2465B/2467B. So, the
PLD will be the same.

I have used the FRAM in 2465/45 A and B model scopes, and in the
early 2465B's that have the battery backup.... which are essentially
simply "A" CPU cards... with a minor change.

I have not used the pull up resistor, though I think I will in the
future.

All the buss drivers are 74HCT parts, so they should all behave
the same as power drops from 5V down to 2V. Zeros on the address
will remain zeros until much lower than 2V. Ones are going to
drift down with the VCC.

If I were writing the sheet, I would not have said may. A data sheet
is no place for weasel words.

Hugo, and I (though independently), spent a lot of time playing
with the FRAM in the 2465B. We have talked about it extensively,
and both ended up wondering what time would tell.

It seems to be ok. If I were really worried about reliability, I
would use an NVRAM. We know it will last 10 years, and *may* last
for more than 30 years.

-Chuck Harris

Mark Litwack wrote:

Hi Chuck,

On the FRAM destructive reads, you're right it's not mentioned in the datasheet. If you read Cypress' papers on how their FRAM works, it talks about it there, and it's also covered in Hugo's paper.

I also agree the FRAM datasheet is very weasely about the power down issues. Using the word "may" in a datasheet is automatically a red flag, at least for me, meaning the bad thing could happen and it's not our problem if it does, or in this case it's not our problem if you didn't listen to us.

You have a good point that if data is getting mangled it would most likely be the first or last address, but again, hopefully. There are also address latches sitting in between to consider.

In digging into that guy's problems on eevblog it got me looking at the compatibility issues. I would have liked to see a few more scope traces while the problem was happening before he installed a pullup resistor and called it done.

And in full disclosure, I don't have a scope that has a DS1225 in it. I was simply interested in why the problem could be happening, which led me to a deeper evaluation of the DS1225 circuitry and its non-compliance with the FRAM datasheet recommendations.

Have you used FRAM in any 2445B scopes? The guy on eevblog had a 2445B which I believe uses a different PLD than the 2465B.

We have a similar heritage starting with 8080A system design in the 80's (or was it late 70s?). I've also been burned multiple times on power-up and power-down issues. Enough to have learned that I don't trust anything out of its operating range even if I think I fully understand the underlying technology. I happily accept a label of overly conservative.

The Dallas NVRAM with its built-in lockout threshold is/was a great solution. It's a shame that Cypress (Ramtron, I guess) didn't duplicate that functionality if they were aiming for the drop-in replacement market.

A minor correction to my previous message: On power up I said the TL7705A waits for the threshold to pass 3.6V before de-asserting RESET. I meant 4.5V.

-mark


On Thu, Jan 21, 2021 at 09:28 AM, Chuck Harris wrote:


Hi Mark,

The TL7705A data sheet specifically states that it will hold
reset/ low when VCC drops below 3.6V, and will keep it low below
2V.

The FRAM data sheet specifically states that holding the CPU in
reset "may be sufficient" to protect the data. I don't like "may",
but for some reason they forgot to say: "won't be sufficient"...
which I think biases their recommendation somewhat. I suspect they
used "may be sufficient" to keep their lawyers happy.

HCT is built identically to HC, except that it is doped to shift
the thresholds down slightly from CMOS's 66%/33%. It was made as a
TTL replacement, so naturally, they would spec it at 5V. Being HCMOS,
its thresholds shift down when VCC drops, just like the HC family,
and It works well below 5V.

The FRAM data sheet specifically states that if CS/ is low, it is
necessary to make sure that WE/ is held high to protect the data
during power down conditions. They don't talk to the destructive
reads... perhaps they forgot?

Probably the most important bit of luck is the 6802 address buss
is not 3-state... but rather composed of TTL compatible output
stages. When the 6802 is driven into reset, the address buss goes
to all zero, or all 1 ... I don't remember anymore. This directs
any accidental diddling of the FRAM to the first, or last memory
byte, neither of which contain the calibration constants or saved
state for the 2465B scope.

I have been designing embedded systems for the last 40 years. My
first was an 8080. I know all the reasons not to rely on the FRAM
in something that wasn't designed specifically to use them... and
when I spec them in my own designs, I follow the manufacturer's data
sheets to the letter. I got burned badly using the first EEPROMS.
The X2816 guys didn't think to mention that power fail would scramble
its brains unless you conditioned the CS/WE lines. DALLAS NVRAMS
saved the day.

The FRAM has been looked at pretty extensively, and FRAMs have been
used in a lot of 2465B's. I don't know what went wrong with the guy's
scope on eevblog. I take everything that happens over there with
a sizable grain of salt. For some reason when there are lots of
pictures, people take that as a seal of authority.

If you are concerned, use a new NVRAM. That is what I always
recommend. I am pretty sure that I won't be around to see any new
NVRAMS fail.

-Chuck Harris




Mark Litwack wrote:
Hi Chuck,

Thanks for the reply and analysis on the /WE line.

I looked at the datasheets for the 74HCT74 and TL7705A (both used in the
2465B):

- The 74HCT series is only specified for 5V +/-10% operation. It's the 74HC
series that's specified for 2V, but that series is not used since it's not TTL
compatible.

- Capacitor C2350 on the TL7705A is a timing capacitor to control the time
on power-up from when the SENSE input passes 3.6V to the de-assertion of
/RESET. It does not keep the TL7705A powered under low voltage conditions.
However, the TL7705A outputs are valid until 2V on power down and undefined
below that. On power up, the outputs are not valid until Vcc reaches 3V (2.5V
typ.).

Even though the TL7705A keeps the processor in reset at lower voltages, and
I agree it's certainly not running, there's still no guarantee the processor
outputs aren't shifting to unexpected levels while Vcc transitions to 0V.
Changes emanating from the processor could reach the FRAM through any of the
other support chips.

More important than /WE, the FRAM datasheet focuses on the /CE line to be
predictably controlled during power down. And that line (pin 20) is
controlled directly by PLDs which also have a 5V +/-10% operating range on the
schematics I've examined.

Even though the /WE line may not be active during power down, there could
still be a spurious read. And with FRAM technology, a read is destructive.
Part of the read cycle is a refresh where the chip internally writes the read
data back to the individual cells (and this is why there is a minimum cycle
time). The cycle must be done while the chip is within specified operating
range on Vdd. Any low on /CE initiates the cycle.

So, with an uncontrolled /CE during power down, the scenario would be that
the FRAM enters a read cycle (or write, depending on signal levels) during the
time where Vdd is below the specified operating range and the chip cannot
reliably complete the refresh cycle.

It's also important to note, unlike the DS1225, the FRAM does *not* have a
specified lockout supply voltage, where it's guaranteed at some voltage to not
perform an operation that would affect stored data. The only thing the
datasheet specifies is that 2.7V is the minimum safe operating voltage, and
for anything below that /CE must track Vdd down to 0V to ensure data
integrity.

One thing I hadn't considered carefully is what happens on power-up.
Similar issues may exist there.

I'm not disputing the FRAMs aren't working. They clearly are. My point is
that it's not sufficiently predictable what's going on below 4.5V when the
FRAM is still active, and without explicitly protecting the FRAM control lines
as per the datasheet it can create a hazard for the stored data.

-mark


On Wed, Jan 20, 2021 at 12:04 AM, Chuck Harris wrote:


Hi Mark,

Thanks for the info.

The 2465b, with NVRAM, has an TL7705A Supply-Voltage supervisor
chip that is in control of the reset line on the processor.

At 3.6V, the chip asserts the RESET/ line, and keeps it there until
the chip times out by way of 15uf capacitor C2350. Because the
capacitor is keeping the supervisor chip powered, it will hold
the reset line at zero until the voltage is well below the 2.7v
safety lock of the FRAM.

So, the processor won't, at least, be trying to write in a bad
power situation.

Further, the WE/ line to the FRAM is controlled by a 74HCT74, U2440B.

The HCT part is guaranteed to operate properly at 2.0V VDD. It
will actually go down to about 1.5V, usually.

Between the two factors, the FRAM should be quite safe... At least
as safe as the NVRAM.

A pull up resistor on the /WE line would be a good idea, but in
practice, it doesn't seem to be necessary.

There was likely something else wrong with the 2465B that was showing
an FRAM failure in the eevblog.

As I said earlier, I have had more than a dozen customers that chose
to have me put FRAM chips in their scopes. None has had any failures.

Life is full of "strokes of luck". That Tektronix used HC logic in
the board was a stroke of luck. That they used a supervisor chip that
was also a stroke of luck. That they didn't use the RAM in a fast
access was also a stroke of luck.

All of these strokes of luck made the FRAM a suitable replacement for
the Dallas NVRAM in the 2465B.

-Chuck Harris


Mark Litwack wrote:
Hi Chuck,

My assertion comes from a series of posts where a user was having issues
with a FM16W08 FRAM, embedded in the gigantic thread over on eevblog:

https://www.eevblog.com/forum/testgear/tektronix-2465b-oscilloscope-teardown/msg2003207/#msg2003207

In short, the issue was that the DS1225 memory support circuitry does
indeed
lockout spurious operation down to the power-down level of the DS1225 as
you
state, but the FRAM remains operational for a good bit beyond that, down to
2.7V. The memory support circuitry does not guarantee any predictable
operation below 4.5V, let alone down to 2.7V.

In addition, the FRAM datasheet is also very emphatic about /CE being held
high when Vdd drops below 2.7V to prevent spurious internal operation.
Again,
not a feature of the DS1225 support circuitry since the designers did not
have
to be worried about it.

The user in the thread was experiencing spurious writes on power-down
because of this, which happened to corrupt one of the cal locations.

I've read Hugo's paper a long time ago, and it spends a large number of
pages on operational timing and FRAM internals, but does not consider the
power-down control signal behavior *required* by the FRAM datasheet.

If you read further in the eevblog thread, the user was able to solve his
issue with a pullup resistor, but speaking from a design perspective, the
memory control circuitry would look different if it was a FRAM in there
instead of a DS1225. In my opinion, as I said previously, it is working by
a
stroke of luck.

And, the 2465B does indeed read and write every address in the NVRAM on
boot. To check that memory is ok, for each location it saves the contents,
writes and verifies two patterns (0x55 and 0xAA), and then restores the
original value. So there is an opportunity for corruption of cal data here
too.

-mark














Re: TEK465 Horizontal trace

wes Bolin
 

I measured C1220 as being good. I also lifted the Neg lead. Slightly more
signal to Q1224 base but still no Left trace.
Q1234 base to ground reads about 24 ohms.

Wes


Re: 2336 on shopgoodwill

Bert Haskins
 

On 1/21/2021 1:57 PM, Bert Haskins wrote:
https://www.shopgoodwill.com/Item/113415082

No affiliation.


It looks nice and clean.

Bert
Sorry it's 2336





2236 on shopgoodwill

Bert Haskins
 

https://www.shopgoodwill.com/Item/113415082

No affiliation.


It looks nice and clean.

Bert


Re: TEK465 Horizontal trace

Dave Peterson
 

Wes,

You said you checked DC values on Q1234. What about the bias circuits around Q1234? E.g.: 37.5v at C1235, -7.4v at C1220?

Also wondering if issues in or around Q1236 might be a problem? If emitter of Q1236 gets pulled down will that pull it's base down and clamp this signal at the base (waveform 80)? Check the DC values around Q1236?

Dave

17241 - 17260 of 193895