Tek 2445A - graticule illumination not working; modify it?

[glossywhite]

Hi all

I have a beautiful Tek 2445A which has just had the famous bad PSU caps replaced, and it was given to me FREE with two "p6137" 10x probes + the full service manual, bound and original (it came from a TV station) - beautiful 'scope! Does anyone know a good source for a new, working graticule illumination board, or do you think it could be modded to allow LED illumination instead, which seems more sensible? I saw somewhere that the board which drives the graticule lamps is very sensitive, and could be blown by not having the exact lamps... any ideas?

many thanks



UPDATE

Oh, GO ON THEN; I'll show my finished 'scope off!

Matt

 

[rumpfy]

Hi Glossy,
There are jokers on this site that would happily kill for a TEK 2445. The good thing about TEK is that they usually provide a full service manual complete with parts lists and circuit diagrams.
The question you ask about the scale lamps is possibly best answered by suggesting that the scale illumination is often achieved by having an incandescent lamp inserted into a hole cutout in the plastic graticule. The light then spreads through the plastic and reflects off any etched lines engraved into the plastic graticule. Changing the lamps to LED types may not work well because you need light to come from the lamp axis perpendicular to the plane of the graticule.
I would further suggest that these lamps are usually driven from a controlled voltage source and the lamps ALMOST NEVER fail. In the case of your new toy, I suggest you get the correct lamps. If the lamp controller board is faulty in some way, I'm sure it would be an easy fix.
Anyway, you are really lucky to have such a beautiful beast. From your other post you say you need to train yourself to get the drivers licence for it. Just study hard and you will be well rewarded.
I have an old TEK 564 and I love it. A mixture of valves and semiconductors and tunnel diodes and an analog storage CRT; Its in the steam radio class but one can do heaps with it.

 

[glossywhite]

It seems the bulb(s) *are* blown, since metering the connector @ J181 header reads open circuit. Strange that you'd design the lamps in series, so that one failure = ALL off...

 

[JimB]

I agree with Rumpfy, keep the incandescent lamps.
They may be old technology, but they are appropriate and correct for this application.

JimB

 

[glossywhite]

I agree with Rumpfy, keep the incandescent lamps.
They may be old technology, but they are appropriate and correct for this application.

JimB

I have to agree actually

Now... to find some...

 

[rumpfy]

Looking at what you showed, my guess is the lamps are 6 volt types. They dont have to be brilliant white in operation. I have some 6 volt and 12 volt miniature types that have no base; only flying leads. These sort of things are used in the dash board of our Toyota land cruiser. They wrap around a plastic moulding. I used some in an aussie built Mitsubishi; these were soldered in to the printed circuit board to show what gear was selected. The lamps are about 12 volt 100 milliamp or less, and the 6 volt are about 120-150 mA from memory. I note the two buffers U130 D and E are in parallel and with a usual TTL gate, the current sinking capability is up to about 30 mA. So with two gates, the capability is 60 mA. I note the pin-out on U130 is not standard TTL, so this might be a TEK special.
In the RS catalog, there is a T1 size lamp with 2 pins for direct soldering. The 5volt lamp is 60 mA. The RS code no is 655-9378.
I'm sure too, TEK will have details in the parts list.
I remember a lamp like this being used in the front panel of the original PDP-8 computers, and they were SENSITIVE.
good luck in your search.

 

[glossywhite]

Looking at what you showed, my guess is the lamps are 6 volt types. They dont have to be brilliant white in operation. I have some 6 volt and 12 volt miniature types that have no base; only flying leads. These sort of things are used in the dash board of our Toyota land cruiser. They wrap around a plastic moulding. I used some in an aussie built Mitsubishi; these were soldered in to the printed circuit board to show what gear was selected. The lamps are about 12 volt 100 milliamp or less, and the 6 volt are about 120-150 mA from memory. I note the two buffers U130 D and E are in parallel and with a usual TTL gate, the current sinking capability is up to about 30 mA. So with two gates, the capability is 60 mA. I note the pin-out on U130 is not standard TTL, so this might be a TEK special.
In the RS catalog, there is a T1 size lamp with 2 pins for direct soldering. The 5volt lamp is 60 mA. The RS code no is 655-9378.
I'm sure too, TEK will have details in the parts list.
I remember a lamp like this being used in the front panel of the original PDP-8 computers, and they were SENSITIVE.
good luck in your search.

Thank you

Also, look what I've found...

http://archive.org/stream/bitsavers_tektronixsvcMay87_38936698/070-6019-01_2465aSvc_May87_djvu.txt

"Scale illumination

The Scale Illumination circuit consists of U13GC,
U130D, U130E, U130F, and associated components. The
circuit enables the operator to adjust the illumination level
of the graticule marks on the crt face plate using the
SCALE ILLUM control.

Components U130C through U130F, depicted on
diagram 4 as inverters, are actually Darlington transistor
pairs. Figure 3-4 is a simplified illustration of the Scale

Darlington transistors U130D and U130E control the
current flow to scale-illumination lamps DS100, DS101,
and DS102. Base drive current for U130D and U130E via
R133 Is set by the front-panel SCALE ILLUM pot R134.
Voltage at the more negative end of the pot is set by the
self-biasing configuration of U130F and R135. The voltage
level established by these two components is two diode
drops above ground (« 1 .2 V) so that, at full counterclock-
wise rotation, the wiper voltage of the SCALE ILLUM pot
will just match the turn-off point of U130D and U130E. The
voltage at the other end of the pot is set by the collectors
of U130D and U130E. As the SCALE ILLUM pot is
advanced, the base drive to U130D and U130E increases,
and the voltage on their collectors moves closer to ground
potential. This increases the current through the scale-
illumination lamps to make them brighter and produces
some negative feedback to the base circuit through the
SCALE ILLUM pot. Negative feedback stabilizes the base
drive to U130D and U130E to hold the illumination level
constant at the selected setting of the SCALE ILLUM con-
trol."


... and at page 3a-16, in document:

http://bee.mif.pg.gda.pl/ciasteczkowypotwor/Tek/2465b 2467b.pdf

 

[JimB]

Looking at the list of replaceable part, Tek are using aged and selected lamps so that the illimination is nice and even.
But I don't think we need get too excited about that.

I am guessing that you are not in the UK, but I found these on the RS Components website, they seem to be just what you need, 5 volt 115 mA wire ended.
https://uk.rs-online.com/web/p/filament-indicator-lamps/0106458/

JimB

 

[glossywhite]

Looking at the list of replaceable part, Tek are using aged and selected lamps so that the illimination is nice and even.
But I don't think we need get too excited about that.

I am guessing that you are not in the UK, but I found these on the RS Components website, they seem to be just what you need, 5 volt 115 mA wire ended.
https://uk.rs-online.com/web/p/filament-indicator-lamps/0106458/

JimB

Hi Jim, thank you for the link. Actually, I live in a little village which is only 11 miles from RS uk HQ in Corby, and one of my best mates is a picker/packer, so this is very convenient

However, for now I may just meter out the illum pcb, find the blown bulb (since they are in series, the likelihood of more than one having blown is small), then I'll make sure there is one bulb on each end to make it even, then temp replace the blown bulb with a 47 ohm resistor.

 

[Inquisitive]

Once you are in there why not replace them all. If they are low cost.

That way you don't have to go in and do it again or a least buy the set to have on hand in the event of the next one fails. Seeing as all this time and effort was spent on researching it.

A ounce of prevention is worth a pound of cure.

Just a thought.

 

[glossywhite]

Once you are in there why not replace them all. If they are low cost.

That way you don't have to go in and do it again or a least buy the set to have on hand in the event of the next one fails. Seeing as all this time and effort was spent on researching it.

A ounce of prevention is worth a pound of cure.

Just a thought.

You are right. However, I have 10lbs of resistors, and 0lbs of bulbs & no money or transport, ergo, resistors win out... for now

Thank you.

PS: Does anyone know if you need to remove all the knobs to remove the front panel, or just the small ones?

 

[JimB]

JimB said:

I am guessing that you are not in the UK

glossywhite replied:

I live in a little village which is only 11 miles from RS uk HQ in Corby

Hmm...
I will award myself a bacon sandwich for electronics and a moldy breadcrust for geography!

JimB

 

[glossywhite]

Okay, so I got the graticule lamp board out, gave the whole area a good clean and metered out the lamps; I was right, only one lamp has blown. However, if I substitute the blown one with a 1W 40 ohm resistor, the working lamps seem a little *too* bright with brightness on full tilt, and they're heating up in excess of 60 deg C, which can't be much good for either them or the light pipe assy. There was a bloke on youtube who'd replaced the blown lamp with only a 10 ohm resistor - I can only imagine how hot it is now, because you can only fit a standard (1/4w?) resistor in the lamp cavity, and there's really nowhere to mount a higher wattage resistor in the area surrounding the assy.

Bulbs it is - I'm a man of precision and like a proper job, not a bodge.

Photos: https://www.flickr.com/photos/22008695@N03/sets/72157640403694524/

 

[rumpfy]

Just now only able to reply and not ignoring other contributions.
I checked your reference which was a ' .TXT ' file type. This had no drawings, so I went for the ' .PDF ' file type and this was HUGE but had EVERYTHING. Its a 37 MByte file but its what you need. By the way, I guess you did not get the original manual. If so, I cannot tell how much you need the original. Just to have the manual AS WELL AS the machine will make you feel king.
The manual makes it all worth while and explains EVERYTHING. Did you note the comment in the manual about how the lamps can be turned off for 'single sweeps' or for photography, or some such. These details of design and manufacture are what you pay for with this stuff and it just makes TEK gear so easy to use. Just TOP engineering.
You have the answers for the lamps and the RS reference is what I was on about in post #6. I couldnt find what I was looking for, but thankfully, Jim B found them. I didnt check on Jb's reference but I'm sure its the goods. I note your comment in post #13; The bulbs are 115 mA and the transistor drivers are probably only capable of sinking 120 mA in total. The lamps have a positive temperature coefficient of resistance so at 115 mA, the lamps are say 45 ohm each but when dimmed to lower currents, currents, the resistance will be higher, much higher. If you have the brightness knob set to a "low" brightness and then replace one of the lamps with say a 39/40 ohm resistor, you may be causing the lamp driver transistors to overheat. So please dont do anything silly; Just get the correct lamps.
The other thing I noted in the manual was that the CRO has a dual timebase. You can not possibly know how useful this feature is when you want to measure pulse risetime on a very low frequency pulse train. The sweep will start at say 10 millisecs per div, and then at the vertical transition of the pulse, the sweep speed will increase to the second stage speed and you will be able to see the transition at say a speed of 10 microseconds per div. Just fabulous stuff. My 564 was new in 1964 and has a dual timebase and I just love it. Nowhere near as good as your new toy but.
I think we are now at the conclusion nes par?
rumpfy

 

[glossywhite]

Thank you for your excellent reply, rumpfy. I actually got the 'scope complete with the HUGE service manual, ring bound in a gigantic black folder with the blue front. The reason I linked to the 2467B service pdf was because, unlike the 2445A service pdf, it is searchable, and since I wanted to find the graticule lamp schematic, and didn't fancy wading through paper at 4am, that's where I found it.

Do the dual timebase features you mention, still exist in my model, or were you referring to the 2467B because I linked to that manual?

You're right; owning the FULL PAPER service manual, tab-indexed, is a great feeling indeed - so blessed I have it.

As for the lamps, yes, they will be properly replaced with the correct type.

 

[rumpfy]

With the dual timebase I would be very surprised if it is not there; BUT; you can make a homework question of it for your self.
In future you'll probably want to use YOUR OWN manual. You can take the manual to bed with you and read it!!!
cheers.

 

[glossywhite]

Each collector of the ULN2003 chip (darlington pairs) can handle 500mA:

https://kamuskomponenelektronika.blogspot.co.uk/2012/12/uln2003.html?m=1


From the 2465b service manual (same driver circuit and IC):

Scale Illumination

The Scale Illumination circuit consists of U130C,
U130D, U130E, U130F, and associated components. The
circuit enables the operator to adjust the illumination level
of the graticule marks on the crt face plate using the
SCALE ILLUM control.

Components U130C through U130F, depicted on
diagram 4 as inverters, are actually Darlington transistor
pairs. Figure 3-4 is a simplified illustration of the Scale
Illumination circuitry, redrawn to show U 130C through
130F as Darlington transistor pairs for the purpose of the
following description.

Darlington transistors U 130D and U 130E control the
current flow to scale-illumination lamps DS 100, DS 101,
and DS102. Base drive current for U130D and U130E via
R133 is set by the front-panel SCALE ILLUM pot R134.
Voltage at the more negative end of the pot is set by the
self-biasing configuration of U130F and R135. The voltage
level established by these two components is two diode
drops above ground (~ 1.2 V) so that, at full counterclockwise rotation, the wiper voltage of the SCALE ILLUM pot
will just match the turn-off point of U130D and U130E. The
voltage at the other end of the pot is set by the collectors
of U130D and U130E. As the SCALE ILLUM pot is
advanced, the base drive to U130D and U130E increases,
and the voltage on their collectors moves closer to ground
potential. This increases the current through the scaleillumination lamps to make them brighter and produces
some negative feedback to the base circuit through the
SCALE ILLUM pot. Negative feedback stabilizes the base
drive to U 130D and U 130E to hold the illumination level
constant at the selected setting of the SCALE ILLUM control.

 

[cowboybob]

Gotta love the HP Engineering department! They've made some really beautiful equipment over the years, but...

For a simple lamp brightness control circuit, this sure looks waaay over-engineered, even to me.

Anyone have a reasonable justification for it?? Just wondering.

(Maybe some very weird MIL spec?)

 

[glossywhite]

Gotta love the HP Engineering department! They've made some really beautiful equipment over the years, but...

For a simple lamp brightness control circuit, this sure looks waaay over-engineered, even to me.

Anyone have a reasonable justification for it?? Just wondering.

(Maybe some very weird MIL spec?)

Yes, that's what I thought too. It's a bit OTT for driving three bulbs; what's wrong with a small transformer winding and a wirewound pot? :S

 

[JimB]

this sure looks waaay over-engineered, even to me.

Ha! That is nothing.
Have a look at this schematic, the cooling fan controller for an HP8640B signal generator.

JimB