Saved a Tek 7503, Tek 455, and BK1420 from storage

[Smokey]

Hey guys.

I recently saved a few scopes from their long exile in storage in the back of an office.  My guess is they probably haven't been powered up in like 15-20 years.

1) The first one is a Tek 7503 mainframe.  I'm kind of excited about this one since I've wanted a 7A22 diff amplifier plugin to play with for a while now, and I finally have the mainframe to use it. 
http://w140.com/tekwiki/wiki/7503
2) The second is a Tek 455. 
http://w140.com/tekwiki/wiki/455
3) The third one is a really odd tiny portable BK Precision 1420.  It is about the size of a modern bench multimeter (like a Keithley 2000) and has a little 2in x 2in display.  This is more of a novelty.
http://www.bkprecision.com/downloads/manuals/en/1420_manual.pdf

So here goes the question since some of you guys are vintage Tek experts......
How do I go about powering these things up considering they haven't seen an electron in 10s of years?  I remember reading somewhere (and it makes sense) that I don't want to just jam the plug in the wall and pray the magic smoke stays in.  Are there caps that are pretty much guaranteed to have gone bad in that period of down time?  Should I ramp up the voltage through a Variac and watch the current?
Anything else I should checkout/replace before powering them up?

[GiskardReventlov]

Should I ramp up the voltage through a Variac and watch the current?

I'm no expert but doing a visual inside couldn't hurt. And I recall reading about the variac idea on these forums once.

Were they at no cost to you? Let us know what happens.

[Smokey]

They look really clean all over.  Only minor plastic damage to the faceplate of the 455.  I didn't see any components that were obviously damaged.

Yup, free to me.

[tautech]

I would use a dim bulb tester, which is just a incandescent bulb in series with the mains.
Say a 100 W bulb. It will flash bright on power up, then should dim if nothing is badly wrong.
It would be wise to check all PSU's for spec and ripple after a visual inspection.
Dust loading on PCB's needs to be checked and cleaned also.

That's the minimum I would do, and there are a few more things, some specific to each scope that others may recommend, Tants, HV resister dividers etc.

[PaulAm]

I wouldn't bother with a variac; a dim bulb tester won't hurt, but it's probably overkill.

The biggest problem with that vintage Tek equipment is filter caps failing open with resultant large amounts of ripple on the supply rails.  This generally screws functionality, but doesn't damage anything.  If something's shorted, you'll get a blown fuse (make sure the correct fuse is in place though)

I just got 3 7603s the other day.  I just plugged them in and turned them on.  They all had issues, but only one seems to have a power supply problem.  All of those had been in storage for at least 5 years.

[tautech]

I wouldn't bother with a variac; a dim bulb tester won't hurt, but it's probably overkill.

Overkill? Its a MUST for ANY unknown old gear until one can be sure it is safe and sound.
Last thing you need is for the magic smoke to escape. :-)

[PaulAm]

In a general sense, I wouldn't disagree with that.  I'm just saying that for Tek gear of that vintage I have not found it to be necessary and I've run multiple dozens of 5000, 7000 and 4xx scopes through my bench.

I would have a greater concern if the dut had a switching ps, but Tek linear PS of that era have common failure modes that do not result in the [further] destruction of the scope if you just turn it on, provided the correct fuses are in place.

Not saying it couldn't possibly happen and if you feel safer using a DB, by all means do so.  Personally, if that 7000 scope has a linear PS, I'd just plug it in and see if it works.

[Seg]

I wouldn't bother with a variac; a dim bulb tester won't hurt, but it's probably overkill.

Overkill? Its a MUST for ANY unknown old gear until one can be sure it is safe and sound.
Last thing you need is for the magic smoke to escape. :-)

Speak for yourself. Some of us like a challenge!

[tautech]

I wouldn't bother with a variac; a dim bulb tester won't hurt, but it's probably overkill.

Overkill? Its a MUST for ANY unknown old gear until one can be sure it is safe and sound.
Last thing you need is for the magic smoke to escape. :-)

Speak for yourself. Some of us like a challenge!

I was really just worried about the OP and his name. Smokey.........

[GiskardReventlov]

Yup, free to me.

Can't beat that price, they look well preserved too.

[Smokey]

The 7503 is alive.  It pulls about 1.6Aac and didn't let out any magic smoke.

Both vertical plugins appear to work correctly, but the horizontal timebase trigger plugin looks a little out of wack.

The trace is only about 1inch wide and the triggering didn't seem to work right.  Now there is a chance I'm just being stupid here and everything is fine, but I ran out of knobs to turn and its still a little tiny trace.  Ill post a picture tomorrow.

[Smokey]

This is what the little trace looks like...  It's just displaying the 60hz pickup.

[David Hess]

The sweep does not appear to be triggered either and the mainframe is unblanked for whatever reason.  It could be a problem in just the 7B53A or the mainframe as well.

[PaulAm]

Could be the H plugin or the H amp. 

Pull the sweep plugin and move the center V amp to the right hand slot.  You now have an X-Y configured scope.  You should be able to move the beam from top to bottom with the position control on the left plugin and right to left with the control on the right plugin.  If you can do that, you have a bad sweep plugin.  If a plugin will move the beam full screen in one slot and not in the other, you have a problem with the associated amplifier.  Be sure the V mode selector is set for the first slot.

Keep the intensity turned down so the beam is just visible so you don't burn the screen.

Also doesn't look like the sweep is being triggered and there may be a visible retrace.  You may need to find another sweep plugin.

Edit: If you can't move the beam to the edge of the screen with either channel, check the PS voltages in the mainframe.

One of the cute tricks you can play with a 7000 series is to put sweep  plugins in both the V and H slots to get a raster display which you can modulate with the Z input.

[David Hess]

Could be the H plugin or the H amp.

The trigger signal from the horizontal plug-in could be missing for whatever reason but the internal trigger path by itself cannot unblank the display so in addition there has to be a problem with the 7B53A or the mainframe.

Keep the intensity turned down so the beam is just visible so you don't burn the screen.

I start by defocusing the beam when doing this if I remember.  This is a good idea whenever fooling with the z-axis circuits to prevent damage to the CRT phosphor.

Also doesn't look like the sweep is being triggered and there may be a visible retrace.  You may need to find another sweep plugin.

The trigger indicator uses a grain of wheat incandescent bulb and could be burned out but the lack of blanking tells me the sweep plug-in is bad.

One of the cute tricks you can play with a 7000 series is to put sweep  plugins in both the V and H slots to get a raster display which you can modulate with the Z input.

The 7000 series is amazingly flexible.  Besides the vertical output, my favorite capability is dual independent sweep and alternate dual sweep.  I do not know of any DSOs which can do that unless you buy two of them.

[Smokey]

UPDATE:

I swapped the 2nd vertical plugin with the horizontal and I was able to sweep the trace full screen in both directions. 
Then when I plugged the horizontal back in, it worked correctly for a little while then went back to the same small trace thing.  So it's intermittent, which is always fun. 

What is more likely, bad connections on the knobs on the front of the plugin or bad connections on the backplane connector?

Another interesting thing is that the intensity knob on the top right corner of the mainframe is a little loose and wiggling it around makes the trace move horizontally.  Not sure what's going on there.

[GreyWoolfe]

I would hazard a guess at the usual suspects.  Make sure the backplane connectors are clean and do the same thing for all the rotary controls for the plugin.  I would also open it up and look to see if there are any flakey components.

[Smokey]

Bad News. 

I got a replacement horizontal plugin, and it does the same thing with the small trace like from the picture upthread.  Looks like I'm going to have to start digging in deeper to the mainframe. 

Add one more thing to the project list I guess........

[David Hess]

What kind of replacement horizontal plug-in did you try?

I am thinking that maybe the load presented to the power supply by the full set of plug-ins including the 7B53A is too high because the input filter capacitors are marginal.  That is easy enough to test by checking the power supply voltages with a full set of plug-ins installed.

[Smokey]

The replacement plugin was a 7B50.

Great idea about the full loading!  That makes sense why the two verticals would work since the middle slot was empty.  I'll try that when I get home.

[David Hess]

The replacement plugin was a 7B50.

Great idea about the full loading!  That makes sense why the two verticals would work since the middle slot was empty.  I'll try that when I get home.

Just install the 7B53A into the middle slot then for a full load test.  Measuring AC ripple on the power supply outputs should be sufficient but I often check the power supply inputs as well after estimating what their ripple should be just to see if the input capacitors have become marginal without measuring them directly which is more difficult.

If the power supply is marginal, then it is not uncommon for some plug-in combinations to present too much load leading to problems.  Some plug-ins are worse than others in this regard and so much so that it is possible to exceeded the rated power supply output and/or the mainframe cooling capacity with some configurations.

With that said, any combination of 7A18s and 7A22s and a 7B53A or 7B50 should work fine in a 7503 despite it being one of the early 7000 mainframes.

How did you end up with a 7B50?  Its 100 MHz replacement, the 7B50A, is much more common as it was produced as a less expensive alternative to the 100 MHz 7B53A for a long time.  The companion to the 7B50, the 7B51, was only produced for a short time and then both were replaced by the 200 MHz 7B70 and 7B71.  One of my 400 MHz mainframes came with a 100 MHz 7B50A which I paired with a 400 MHz 7B85.

[Smokey]

Full disclosure: I don't know anything really about tek mainframes.

I just found a horizontal plugin on ebay that was local and drove over and picked it up for 25 bucks.

[David Hess]

Well, assuming the 7B50 works, there is no reason to prefer the 7B50A to it.  The 7B50 is just older.

Besides mechanical differences, the 7B50 uses tunnel diodes for triggering probably making it higher performance, has a somewhat odd combined level/slope control which was shared by the 7B51, 7B70, and 7B71, and includes XY mode standard which is optional on the 7B50A and the 7B50A only supports XY mode from an internal source anyway.  This means you can use the 7CT1N curve tracer with the 7B50 (at least theoretically) but not with the 7B50A.

I kind of wish there had been a 7B51A to go with the 7B50A but understand why there was not.

[Smokey]

Removing the middle vertical plugin made the horizontal kind of work better for a while.  The trace covered the whole screen like it's supposed to, but the triggering didn't seem to want to work right.  Then after a while it went back to having the same small trace thing like when all three plugins were installed. 

Which power supply voltages were you talking about checking ripple on?  I checked some of the regulated voltage test points (+15, -15, etc) that were easy to get to and they all looked good.  That mainframe is seriously intimidating to take apart with all the tied wire harnesses running all over the place.  It looks like one of those things that I'll never get back together correctly if I start to take it apart.  And it looks like you have to dig pretty deep to get to the AC power supply boards.

[David Hess]

Removing the middle vertical plugin made the horizontal kind of work better for a while.  The trace covered the whole screen like it's supposed to, but the triggering didn't seem to want to work right.  Then after a while it went back to having the same small trace thing like when all three plugins were installed. 

Which power supply voltages were you talking about checking ripple on?  I checked some of the regulated voltage test points (+15, -15, etc) that were easy to get to and they all looked good.  That mainframe is seriously intimidating to take apart with all the tied wire harnesses running all over the place.  It looks like one of those things that I'll never get back together correctly if I start to take it apart.  And it looks like you have to dig pretty deep to get to the AC power supply boards.

I do not have complete 7503 documentation but I do have documentation for the 7403 which should be practically identical.  The 76x3 series is very similar as well.

There are a set of test points on the right side toward the rear as you face the front for the +/-50 volt, +/-15 volt, +5 volt, and +130 volt supply outputs shown on page 5-14 of the linked PDF.  From your description, at least some of the DC outputs are sagging but I would expect that to be accompanied by significant ripple.  The -50 volt supply is the reference for all of the others so if it is low, none of the outputs will be correct.

I am including cleaned up schematics that I made for the 7503 in PDF form as attachments (*) and here is a link to the 7403 documentation as well.

http://bama.edebris.com/manuals/tek/7403n/

If you know where full documentation for the 7503 is, I would be interested.

(*) The misbegotten forum software puked on the attachments and destroyed my post in the process so they are not included.