7503


donlcramer@...
 

After reading Stan's post, I went to the Tek web site on 7K history to see
what a 7503 was. There it said the 7503, 7504 and 7514 were only about 1
year in production.

So I am wondering, were they too close the the 7603 in performance, or was
there some other reason they were pulled so early?


dhuster@...
 

The 7500-series were some of the first off the 7K product line and
were riddled with problems. It's not unlike the 576 curve tracer --
looked "modern" on the outside, but inside used 500-series wiring and
components. The 7500's were the guinea pigs for a totally new scope
concept and a lot of ideas that seemed really neat at first were
abandoned with the newer designs. Those 237 little incandescent
lamps in the 7A12 were cute the way they lit up the little buttons,
but it didn't take them long to figure out that it was a maintenance
nightmare.

Dean


Miroslav Pokorni <mpokorni@...>
 

Did you mean 237 individual lamps? Things that I opened (few plug-ins and a
military version of 7603) had one lamp per switch assembly, i.e. one lamp
provided light and clever positioning of depressed button facilitated light
piping. I am sorry to say that I noticed that Tektronix was buying preburned
Japanese brand small lamps. I fully understand Tektronix motivation for
that, I am just sorry that reasons seem to be real. Personally, I would buy
a GE lamp only under duress, small, household or car lamp; it is my
impression that ever since GE moved lamp manufacturing out of Nella Park,
quality went down the tubes.

But, back to Tektronix, 237 lamps would really put a crimp on MTBF. Out of
all 7000 Series devices that I handled which contained lamps, probably a
dozen of plug-ins and a half a dozen frames, I have seen only a couple of
failed lamps and I got them that way. I am sure that everyone would be
horrified to hear, but I bought replacements from local Radio Shack.



Regards

Miroslav Pokorni

-----Original Message-----
From: dhuster@pb.k12.mo.us [mailto:dhuster@pb.k12.mo.us]
Sent: Friday, October 12, 2001 5:47 AM
To: TekScopes@yahoogroups.com
Subject: [TekScopes] Re: 7503

The 7500-series were some of the first off the 7K product
line and
were riddled with problems. It's not unlike the 576 curve
tracer --
looked "modern" on the outside, but inside used 500-series
wiring and
components. The 7500's were the guinea pigs for a totally
new scope
concept and a lot of ideas that seemed really neat at first
were
abandoned with the newer designs. Those 237 little
incandescent
lamps in the 7A12 were cute the way they lit up the little
buttons,
but it didn't take them long to figure out that it was a
maintenance
nightmare.

Dean



------------------------ Yahoo! Groups Sponsor

To unsubscribe from this group, send an email to:
TekScopes-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/


dhuster@...
 

I was just taking a little license there with the 237 lamps. The
7A12 was lousy with relays and they were the worst problem. I'll
take a 7A18 or 7A26 over a 7A12 any day. But I've replaced billions -
- another exaggeration -- millions of incandescent lamps in Tek
gear. The pointless pilot lamps in the TM500 function generators,
knob skirt lamps in the 465 & 475 and worst of all, the knob skirt
lamps in the 434. The 485 was ahead of its time, using LEDs around
the knob skirts rather than incandescent lamps under them. The 5000-
series used neon lamps still! The 7000-series put it all on the
CRT. The new DSOs don't worry about it. They just calculate the
values and any idiot can just read it off the display -- and be right
if the scope isn't using the wrong calculation points!

Dean


Miroslav Pokorni <mpokorni@...>
 

Dean:

Tell us about these relays; I think it was you who referred to them as
'white relays'. What kind of relays they were? Encapsulated reed, I would
think. What kind of failures did you experience with them? Were coils
failing or contacts did not do well in these super dry circuits. It is a bit
unexpected how coils in reed relays can be fragile. I recently had a failure
of a PRMA relay (14 pin DIP package); relay was used in a car and driven by
a switch with an additional load besides relay itself. I did not bother to
put a suppression circuit (diode or resistor, as they like to use in cars in
these days). After a year of low hour use, relay failed and failure was an
open coil. I guess, between spikes and 13 V running voltage, winding was
fried.

The 7000 has a fair share of incandescents, both, in the mainframe and
plug-ins. If specs in the manual are to be trusted, these incandescents are
burned in and selected for light output, what is some assurance of life
expectation, too. At the time of 7000 Series introduction the LEDs were
quite new, somewhat dim, power hungry and short on reliability; drive
current was in excess of 20 mA and die mount just did not seem to be able
to hack these miserly 40 mW or so. And dazzling color choice should not be
forgotten either, anything you want as long as it is red. So, choice of
incandescent lamps is not quite surprising.


Regards

Miroslav Pokorni

-----Original Message-----
From: dhuster@pb.k12.mo.us [mailto:dhuster@pb.k12.mo.us]
Sent: Monday, October 15, 2001 10:23 AM
To: TekScopes@yahoogroups.com
Subject: [TekScopes] Re: 7503

I was just taking a little license there with the 237 lamps.
The
7A12 was lousy with relays and they were the worst problem.
I'll
take a 7A18 or 7A26 over a 7A12 any day. But I've replaced
billions -
- another exaggeration -- millions of incandescent lamps in
Tek
gear. The pointless pilot lamps in the TM500 function
generators,
knob skirt lamps in the 465 & 475 and worst of all, the knob
skirt
lamps in the 434. The 485 was ahead of its time, using LEDs
around
the knob skirts rather than incandescent lamps under them.
The 5000-
series used neon lamps still! The 7000-series put it all on
the
CRT. The new DSOs don't worry about it. They just
calculate the
values and any idiot can just read it off the display -- and
be right
if the scope isn't using the wrong calculation points!

Dean



------------------------ Yahoo! Groups Sponsor

To unsubscribe from this group, send an email to:
TekScopes-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/


dhuster@...
 

Miroslav,

I wasn't the white relay guy. But as far as test and burn-in, it may
be that Tek wanted all the lamps to have the same light output so
that those vertical and horizontal mode switches across the front of
the 4-holers would be the same intensity. And I suppose that they
did the burn-in figuring that most lamps, if having a manufacturing
defect, will fail within the first hours of life.

Dean


Miroslav Pokorni <mpokorni@...>
 

Dean,

As you said, the measurement after burn in is mostly to equalize light
output. Additionally, if intensity sticks out of expected range that is an
indication of a manufacturing defect. All these lamps in 7000 Series are DC
powered and there is an effect called 'filament notching' that makes a notch
in tungsten filament and shortens life. I do not know how that notching
comes about, I would have to look it up.

I found out about 'filament notching' in a Chicago Miniature application
note, which I do not have here. In these days of Internet you would say go
to Chicago Miniature web site, but there are no application notes there, so
I would have to look it up when I get home.


Regards

Miroslav Pokorni

-----Original Message-----
From: dhuster@pb.k12.mo.us [mailto:dhuster@pb.k12.mo.us]
Sent: Wednesday, October 17, 2001 12:59 PM
To: TekScopes@yahoogroups.com
Subject: [TekScopes] Re: 7503

Miroslav,

I wasn't the white relay guy. But as far as test and
burn-in, it may
be that Tek wanted all the lamps to have the same light
output so
that those vertical and horizontal mode switches across the
front of
the 4-holers would be the same intensity. And I suppose
that they
did the burn-in figuring that most lamps, if having a
manufacturing
defect, will fail within the first hours of life.

Dean



To unsubscribe from this group, send an email to:
TekScopes-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/


Michael Dunn <mdunn@...>
 

At 1:51 PM -0700 2001/10/17, Miroslav Pokorni wrote:
powered and there is an effect called 'filament notching' that makes a notch
in tungsten filament and shortens life. I do not know how that notching
comes about, I would have to look it up.
As I recall, a notch starts out due to filament evaporation (in a random spot, or one already slightly thinner than average). This thinning creates a hot-spot, which only accelerates evaporation from that spot. I think it can happen with AC too. It's what halogen lamps try to reduce, by depositing the evaporated tungsten back on the filament. I think.

Michael


John Rehwinkel <spam@...>
 

At 1:51 PM -0700 2001/10/17, Miroslav Pokorni wrote:
powered and there is an effect called 'filament notching' that makes
a notch in tungsten filament and shortens life. I do not know how
that notching comes about, I would have to look it up.
As I recall, a notch starts out due to filament evaporation (in
a random spot, or one already slightly thinner than average).
This thinning creates a hot-spot, which only accelerates evaporation
from that spot. I think it can happen with AC too. It's what
halogen lamps try to reduce, by depositing the evaporated tungsten back
on the filament. I think.
Actually, that's the mechanism that takes over once the notches
appear (creating hot spots). The notches themselves are in fact
caused by tungsten atoms migrating in the direction of current
flow along the fault lines between grains. In AC use, this only
tends to occur where there's a thermal gradient (i.e. the ends
of the filament). In DC filaments, this occurs along the entire
unit.

A web search on "filament notching" yields lots of hits,
including these two (both have illustrations). The second
one even mentions that filaments drawing less than 40mA
(i.e. thin filaments on low candlepower lamps) at 1800-2250K
(typical long life lamp temperatures) are particularly
susceptible.

http://www.htl.co.jp/pro/kogata/tokusei_e.html
http://www.lumex.com/tech_notes/mini_txt3.html

-- John Rehwinkel KG4L
spam@fgm.com


dhuster@...
 

Miroslav,

I was always under the impression that the whole thing was caused by
the Edison effect, metal particles being attracted by a surface
external to the lamp, charged positive with respect to one side of
the filament, the metal particles finally darkening the inside of the
envelope killing light output. Eventually, enough metal erodes from
the filament that a low-wattage capability section of the filament
develops that can't handle the inrush current at turn-on, gets a
little too hot and melts the filament into an open.

Steve Schmelzer, a former Tek employee with whom I served in the U.S.
Navy, used to quip that a lamp operates not put emitting light, but
bu sucking up dark. They fail when they've sucked up all the dark
that they can hold, obviously proven by the dark insides of the
lamp. And such is how many theories are born.

Dean


mwcpc7@...
 

In a message dated 10/17/2001 6:02:07 PM Eastern Daylight Time, spam@fgm.com
writes:

In DC filaments, this occurs along the entire
unit.

A web search on "filament notching" yields lots of hits,
including these two (both have illustrations). The second
one even mentions that filaments drawing less than 40mA
(i.e. thin filaments on low candlepower lamps) at 1800-2250K
(typical long life lamp temperatures) are particularly
susceptible.
Maybe this explains why putting a diode in series with night light lamps
(which I've done for years) doesn't extend their life any where near much as
I thought it should.

Mike Csontos


Don Black <jeans@...>
 

This raises another thought. When an AA5 radio needs a replacement ballast
resistor (for whatever reason), one trick is to place a diode in series with the
heater supply to reduce the effective voltage to 0.707 of the supply.
It forms a lossless half wave rectifier and since power is only supplied for
half a cycle the average power is also halved. Because of the voltage squared
power law, the effective voltage is the square root, not half.
Anyway, it's an effective method, especially in small sets where the reduced
heat is important but I wonder if the DC reduces the tube life.
The heater isn't as hot as a lamp filament and it's coated with an insulating
slip that might protect it. Has anyone used this technique and noticed any
increase in tube heater failure?
Don Black.

mwcpc7@aol.com wrote:

In a message dated 10/17/2001 6:02:07 PM Eastern Daylight Time, spam@fgm.com
writes:

In DC filaments, this occurs along the entire
unit.

A web search on "filament notching" yields lots of hits,
including these two (both have illustrations). The second
one even mentions that filaments drawing less than 40mA
(i.e. thin filaments on low candlepower lamps) at 1800-2250K
(typical long life lamp temperatures) are particularly
susceptible.
Maybe this explains why putting a diode in series with night light lamps
(which I've done for years) doesn't extend their life any where near much as
I thought it should.

Mike Csontos


To unsubscribe from this group, send an email to:
TekScopes-unsubscribe@egroups.com



Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/