Tektronix Type 547 No trace

[Zalgar]

Hello!

I was wondering if someone would be able to give me some pointers on this old scope I picked up, its a Tektronix type 547 with type 1A1 plugin.
Im unable to get a trace on the screen, all I get is a trace on startup (see youtube link below) and after the trace appears on the screen, I get a single glow in the centre.
I've played with all the settings and all I can get it to do is flash the glow in the centre, there's also a link to the image of the aforementioned below.

Thanks for any help!

[Alex Eisenhut]

I didn't look at the video, but you have a "trace off-screen" light lit up over the CRT at the left. That right there is all you need to know.

The glow is probably from secondary electron bombardment as the electron gun is firing towards the Sun instead of the screen. (Turn off the scope, you are ruining the CRT electron gun. Save those remaining hours!)

Usually this is caused by a faulty plugin, a shorted semiconductor that causes an amplifier to output rail voltage somewhere inside the signal chain.

Spend some time reading the excellent service manuals, with the scope nicely OFF...

Also be aware the 547 has a design flaw in the high voltage section; after about half an hour of running, the HV loses regulation and the CRT goes "soft" and loses all focus and brightness.

You can have more than one problem at a time!

The solution for the HV flaw, unfortunately, requires disassembly of the HV section, removal of the transformer, and having it re-potted and rewound by an expert. This is going to be difficult as the people who know how to do this are getting rarer... 

I had this done years ago by Bill Schell. No, I have no idea how to do it myself, it's a technical skill that requires practice!

Here is a place to start looking

http://www.reprise.com/host/tektronix/reference/547_hiv.asp

Keep in mind this information is quite old.

Also check out

https://groups.yahoo.com/neo/groups/TekScopes/conversations/messages

and

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

[Alex Eisenhut]

Meh, it's not really a "design flaw", I mean the thing worked perfectly as designed for decades; it's more of a material problem, after that much time the potting compound changed properties and the HV section can't work properly.

What you can do is run the 547 without the plugin and see if you can center the CRT's trace. There is a dummy load resistor in the scope so it should power up without a plugin and work normally.

There is a switch for this resistor in the plugin bay, on the top right hand side. It actually has three positions, it is pushed in by the plugin, pops out with the plugin removed, and you can also pull on it manually and it will lock in a "more out" position. Push it back to make sure it is in the "neutral" position so you are sure that resistor is in the circuit. You never know, the spring might be weak.

This switch allows using the plugin on an extender without the extra load of the dummy load resistor. Something you might end up doing if you get a plugin extender.

Did you play with the "trace separation" knob under the CRT? Keep it at 0 for now.

[Zalgar]

Thanks for the replies!

Under your recommendation I've tried to centre the trace without the plugin, same deal, though this time I've only got a horizontal trace off screen indicator.
Im stumped!

Trace separation is at 0.

You seem to have a lot of experience with this scope, do you have one?

[Alex Eisenhut]

Yes, I have a 547. It came from Radio Canada, the federal broadcaster up here. What's the history of your 547?

Well, let's see; removing the plugin removes the vertical off-screen light, OK, not bad, did you play with the horizontal position control? You see, sometimes the horizontal off-center light also just means "time base armed, waiting for trigger".

Put the horizontal display control to A. Put all the levers UP. Keep the time/cm control at something like 1 msec/cm. Keep the intensity at around 4. This scope is capable of enough intensity to permanently burn the CRT if you let it. As a matter of fact, keep it out of focus so you don't generate an intense spot.

You should be able to generate a trace by fiddling with triggering level and the horizontal position knob.

If not, I'd check the tubes. With the power OFF, remove the side panels, use a coin to turn the big screws holding the panels, you'll find on the scope's right hand side (as seen from the front) there is another massive frame of electronics, it is held in place by two thumbscrews. Open them and pry the frame open, it swings out. It is not necessary for basic operation; this frame is the B time base, so with the horizontal display set to A you know that the open frame is not adding to the confusion.

Now power on the scope and turn off the room's lights. Now you can check for tubes with a dead filament. Don't laugh, I am lazy and fixed a 531 this way once. Simple!

Of course, for now resist the temptation to fool around in there while it's on; what doesn't burn you will probably zap you. It's dangerous in there. Getting a DC shock is much more unpleasant than AC. (And there are fragile parts in there, just brushing against a tunnel diode may be fatal for it, and these parts are tough to source.)

Once you know that all the tubes are lighting, then you can move on ... If not, well, you need to shop for some tubes first. Keep in mind these are signal tubes, they light up dim reddish orange to orangey. They don't light up like lightbulbs. The only "tube" that doesn't light up is the delay relay underneath the chassis. I assume the scope does its normal power-up sequence, it takes about 20 seconds after power on to start the high voltage supplies. These rarely fail.

So, how do you feel about probing high voltages with this thing on?

PS: Don't leave it on too long with the covers off, the covers need to be closed to properly cool the thing. Especially applies to the plugin, it will have almost no airflow if you remove the covers of the scope.

[Zalgar]

I just recently picked up this scope, apparently it was from an old engineer who worked a Tek.

Still unable to get a trace, all but two tubes aren't lit. Are they meant to all be on regardless of settings?
One that isn't lit is on the left hand panel closest to the plugin module and the other is on the right hand side next to the two large cylinders (presumably tubes encased in a metal shroud) Sorry I'm not alot of help at helping myself Lol.

From what I can see they are the only tubes not working or at least glowing, is there a good place to get tubes other than the one and only Ebay?

Now in regards to poking around inside, I'm relatively confident I won't kill myself... What am I looking for?

Cheers

[Alex Eisenhut]

The filament is always powered no matter what. It's the first thing to power up when you turn it on, (and the fan), then the high voltages kick in when the delay relay turns on. This is good for the tubes, makes them last longer.

"One that isn't lit is on the left hand panel closest to the plugin module"

Probably the 12AT7. The fun thing is the part number is printed on the chassis.

"the other is on the right hand side next to the two large cylinders (presumably tubes encased in a metal shroud)"

Well, that sounds bad. Is it the big fat tube or one of the tall skinny ones? The two large metal cylinders are capacitors, leave them be for now. Plenty to worry about already.

If a tube is dead in the power supply, there might be unregulated voltages being sent all over, so I'd keep the scope OFF until you can get new tubes.

(Hmm, looks like I was wrong, the delay relay is in fact on the top of the chassis, in a metal case. I must have been thinking of the 531. Or simply dreaming.)

Try to get in touch with the seller, there may be a bunch of tubes there. That's the best case.

Right now, trawling eBay for a good deal on tubes is a nightmare. Prices seem really high and the selection ... not impressive. One problem is that Tektronix liked to pick the best tubes out of a batch and put their own part number on it. While you can use just regular tubes, you might lose some bandwidth or flatness or something. Does it really matter?

In the meantime another thing to check is the rectifier diodes. On the bottom of the scope there is a panel that comes off, with the power off and unplugged, check the diodes. AFAIK 547's all had silicon diodes.

Some good news, seems in the meantime someone has figured out a way to fix the HV problem without rewinding the whole thing.

https://richardsears.wordpress.com/2015/08/23/tektronix-547/

First get the thing to run, then see if it also has the HV problem. If the previous owner already fixed this, you're OK.

"Now in regards to poking around inside, I'm relatively confident I won't kill myself... What am I looking for?"

Dead mice. Bugs. Seriously. Until you can get some tubes, there's not much to do, read the theory of operation section of the 547 manual. Ignore the plugin for now, without good power going to it, and without a proper signal chain in the scope, it's not doing anything.

Just eyeball the thing and try to find circuits on the diagrams. Because there is no PCB, you are limited to finding the refdes of the active parts only, since they are printed on the chassis. It's a different approach than a silkscreened PCB! This will help you when troubleshooting.

[Zalgar]

Oh wow, Didn't notice the part numbers on the chassis, I should have expected as much from that era!

You are correct its the 12AT7, the other one next to the caps is 6N030T

I bought the scope off a guy who is local, he sold it as working and did show it powering up (No trace then, should've been more attentive...), so either hes full of it or It was damaged during transit, I bought it for relatively cheap $80AUD.

As for replacement tubes, I've found a couple lots on ebay, do you know of a way to check compatibility of replacement tubes? First time working with valves, excuse my ignorance...

I doubt the previous owner had fixed the HV issue, but here's hoping. The process seems relatively simple, I will just have to source components. That being said the expenses of this is quickly piling up... I might just restore it as a preservation of old tech, plus no better way to learn about valves.

The diodes are all silicon below, and are all good from my tests, I'm now just poking around checking for any dead resistors, loose wires etc.

[Alex Eisenhut]

Bah, the 6N030T is the delay relay. This is a bit confusing, in my 547 it is in a metal sleeve. This tube doesn't light up since it isn't really a "tube", it's a thermostat that closes a contact. It doesn't rely on a hot filament to work, just a warm filament to warm itself up. Basically this is OK then. If all the tubes are lit on the power supply side, the voltages are much more likely to be correct then. So no problem turning on the scope!

Well, you can wiggle the 12AT7 out of its socket and check the pins for dirt. Try to eliminate the socket and the wiring as the cause. You simply can never know exactly what the previous owner did so always assume the worst.  Also, does this tube have a lovely silvery mirrory spot on it? Often seen on top?

12AT7 is the part number you want, however Tektronix sold them with their own Tek part number, these are likely to be very expensive, and may simply not be that important for now. Does that 12AT7 have a Tek sticker on it? If not, you know it's already been replaced with a generic 12AT7. Or the label fell off! 

Used tubes are fine as long as the getter, that mirrory silvery thing, is big and bright and shiny. Black discoloration indicates overheating. Might still work, but it has a lot of hours on it already.

A tube that has lost its "getter" completely (clear glass) has lost its vacuum and will not work. Of course, the delay relay never had a getter in the first place so it's fine.

[Zalgar]

The only tube that's in a metal sleeve is the 12AU7, at least in mine.
I wiggled the 12AT7 out, its an AWA Tube, so evidently it has been replaced, it still has a silver top no black or anything.

From the looks of it, there are a few on ebay, but they are fairly expensive, but manageable. I have also seen some on guitar sites, for amps.

It doesn't particularly matter if I lose some bandwidth, I just sort of want to get it running. I tested the tube with my multimeter, all the pins are doing what they're supposed to without checking with a proper valve tester I cant be sure.

[Martin.M]

in my little community is also somthing to find about 547`s 
see http://www.wellenkino.de/forum/thread.php?board=1&thread=214&page=1
something about the restoration of old Tek scopes: https://www.eevblog.com/forum/testgear/vintage-tek-restoration-pictures-by-martin/

before you start to repair anything:
1.) the Tek have to be clean 
2.) checking all Voltages of the low voltage PSU; (Manual, from page 3-3 and following)

at first:  -150V  (this is the referece for all other voltages and must be accurate <1V )
then +100V,  +225V, +350V*
check them all also for a ripple (Scope, ac coupling)

If there is anything wrong your 547 will generate errors allover, good food is important to do a great job 

not lightning tubes in the plugin and elsewhere: Some of the tubes are serial heated, connected to the +100V. If one of them strikes all of them goes dark, means: a dark tube must not be defect also.

* it may be possible that +350V is a +125V in series with the +225, so the electrolytic there is at the case elevated +225V, not to the ground. Be carefully and good isolated

greetings
Martin

[Alex Eisenhut]

Hey I think the 12AT7 must be good, since removing the plugin changed the vertical position. If the 12AT7 was dead, the entire vertical path would be dead either way.

Test the filament for resistance, if it's a few ohms, should be good.

Definitely get ready to test voltages.

Also, be aware that soldering to the ceramic strips should be done carefully (read the manual), and you should use 3% silver-bearing solder. There might be a spool of solder left inside the scope. You can still get 2% silver-bearing solder, it's good enough.

[Zalgar]

Spent the day giving the scope a good clean, tubes were cleaned as well. It looks so much better lol.

The tube tested 6 ohms on the filament, so from the looks of it its good. Could I pull a non essential tube and stick it in place to see if it is the issue? It looks almost brand new...
After work tonight I'll get at voltage testing, I've also been scouring ebay for replacements, which seems to be a winner either way.

Just as you mentioned inside the case tells me to use silver-bearing solder, will have to acquire some of that as well.

Thanks for the help guys, I would be getting absolutely no where without these pointers 
 
edit: I've found another tube which isnt lit, labeled v91 which is a 12AUG type.

[Alex Eisenhut]

Hm, good, good. It really pays to check the simple stuff, with no filament power the electrons don't come out to play and any troubleshooting is useless until you get working tubes.

Clean tubes are better, clean them carefully with dry microcloth, some tubes had water-soluble markings like Amperex Bugle Boys. If anything, you can sell Bugle Boys for a premium. Clean tubes let the glass run cooler. Ironically, even though tubes need heat to work, the cooler the glass runs, the longer it will last.

That tube is probably a 12AU6. In general, the first set of digits is the nominal filament voltage, 6 =6.3V and 12 = 12.6V. You'll mostly see 12 and 6 in a scope. TVs had more varied voltages. Rectifier tubes where usually 5V (I forget why specifically), but the 547 uses silicon diodes, and special HV rectifier tubes with ~1V filaments. But you won't see those unless taking apart the HV section. Don't mess with those just yet, the filament is at high voltage relative to ground. = big ouchie.

There is also a cold cathode tube with no filament, it's like a huge neon bulb really, in the 547 it is used as a voltage reference, this one you'll only see light up when the DC power kicks in after the power-on delay. This is normal. It also has an enormous getter that covers a lot more of the tube.

The letters are just some general type, and the last digit is active elements + filament. So 12AT7 kinda means 12V filament, and 6 elements which means dual triode (3+3+1). And guess what? 0 means no filament, and that cold cathode tube is ... 0C3.

12AU6 is a pentode (cathode, grid, screen, suppressor, plate + 1)

Then you have tubes with just 4 digits which are often industrial versions of some tubes, or just plain old industrial. For example, the 547 uses 6080 tubes, which are the big monsters in the power supply section, these are dual triodes.

The 12AU6 has an industrial version called the 8426.

Then there are the European-style part numbers. ECC88 for example is a version of the 6DJ8. Also, they liked switching around the letters and numbers to indicate special versions.

And then there's the whole spectrum of Russian tubes, good tubes, but they used Cyrillic characters and I'm not too familiar with them.

Except for... Nuvistors, which may look like transistors to you because of their small size and metal case, but are actually tubes. There aren't any in the 547... but there might be in the plugin! The early 1A1 uses them. Those I have Russian versions for and they work fine. Also in the 1A1 they are safely hidden in a weird metal contraption on rubber mounts, since Nuvistors pick up mechanical energy and turn it into electrical signals, called "microphony".

Early 1A1s have the DC-GND-AC input selector as a big ring around the BNC, later 1A1s have a regular switch. Late 1A1s have FET inputs. BUT, early 1A1s may have been fitted with FETs by the user. Only way to tell is to look. But in all versions, there are Nuvistors in the output section.

Nuvistors still show up by the boatload on eBay. Not too bad.

https://en.wikipedia.org/wiki/RETMA_tube_designation

Why am I dumping all this on you? Well, the more you know the more tricky you can get with finding parts. Some people just get locked into one part number, but if you search for substitutes you can find deals!

http://www.worldtubecompany.com/mm5/merchant.mvc?Screen=TSL

Also, sometimes you'll see something like a 6AU6, it's a 6V version of the 12AU6. Worst case, you can hack that into service but don't do that just yet. Tubes were made by the quadzillions back then and there should still be plenty.

Hey, then later we can get to the transistors and tunnel diodes! And drifting carbon composition resistors!

Isn't this fun? And you haven't even looked at a single waveform yet!

And yes, you can slap in equivalent tubes from elsewhere, but that's just borrowing trouble right now.

[Zalgar]

Thanks for the really indepth replies! Im learning so much, and as you said I haven't even looked at a waveform
You are correct, it is a 12AU6, brain didn't click when I wrote that down.
Okay, so from this point I need to replace the 12AU6 which isn't lit, it was located on the panel which was located on the frame which swings out, though from what you said, it shouldn't be relevant to running the scope.
Aside from that I'm working on testing the high voltage rails, what are some good indicators of it being the problem?

[Martin.M]

your checklist have to start with:

-150V  =?
+100V =?
+225V =?
+350V =?

If there is a problem, that is to repair at first. Then continue 

greetings
Martin

[Alex Eisenhut]

your checklist have to start with:

-150V  =?
+100V =?
+225V =?
+350V =?

If there is a problem, that is to repair at first. Then continue 

greetings
Martin

Excellent advice. The -150V I believe is the reference for the rest, if that one is out, the rest can't be good either.

Zalgar, did you know your heart is roughly the size of your clenched fist? Now look at the size of the power transformer in the 547. It's like The Terminator, it feels no pain, or fear, or pity, or remorse...

Be careful in there.

You can test voltages with the plugin removed. At least this isolates faults and protects the 1A1 in case of power supply failure while you tool around in there.

Here's another good read for the 547

http://bama.edebris.com/manuals/tek/547notes/

[Zalgar]

I'm gonna sound like an absolute idiot, but I cannot seem to find the test points for the HV rails, I do see the -150v pot, but I can't trace it back to the -150v rail...

I also cannot find any reference to testing the voltages in the manual, only the specs of the voltages.

[alsetalokin4017]

Tektronix typically puts voltage values at various places on the schematic, which can be found in the Instruction Manual for the 547 at the "bama" link that Alex posted up above. You test these voltages by finding the locations using the board diagrams in the manual, using a DMM with the negative probe lead clipped to a solid chassis ground, and the positive probe lead to the desired point in the circuit. Or, in case of the primary transformer winding outputs, connect the DMM across the transformer output terminals themselves.  See the schematic section in the image below. You can see several voltage test points with their normal voltages indicated in various places on the schematic. The block diagrams are also useful for tracing voltages and waveform test points.

Please forgive me if this sounds too basic. Maybe I'm not understanding your post.

Note that several of the filament supplies are lifted to high voltages wrt chassis ground. So be very careful!

On an old Tek RM503 scope that I have, I eventually found that the HV elevation of the nominally 6 volt CRT filament supply was "leaking" inside the main power transformer and was causing the "lack of HV and fuse blowing" problem I had.

[alsetalokin4017]

Also... there are some weird features of the 547's filament supplies, that I'm seeing on the schematic. It looks like most of the 12 volt tubes also have their filaments in simple parallel with the 6 volt tubes! The exceptions being the 12AU6 V91 and the 12AT7 V1003. These two tubes may not light up if the plug-in isn't in place.... I think. See the section of the PS schematic I attached above, and the filament schematic attached below.

Check the filament of the 12AU6 for resistance before you decide it's bad simply because it isn't lighting up. Also you can check the voltages at the tube socket to see if it is getting any filament voltage to begin with.

[Zalgar]

Im aware of basic voltage testing, Its a very complicated device didn't know where to start

Right so I got for:
-150V  = -152.7
+100V = 98.8
+225V =228.5
+350V =357.0

I adjusted the -150 pot and got:
-150V  =-150.0
+100V = 101.4
+225V = 224.4
+350V = 350.7

Winner Winner Chicken Dinner! She's got a beautiful crisp line! So happy!

However its only time base A of which that is working, could that not lit tube be the possible reason for B not working? I'm getting a horizontal deflection indicator for it aswell which even with the finest of fiddling of the horizontal position knob does not bring it to centre. The tube is the 12AU6 that I mentioned before, I've got 11 ohms on pins 3 and 4 and no continuity between other pins.

Getting A running has given me a second wind! Getting really well into it now...

edit: I plugged the plugin back in, the tube is now lit and both bases work, Im feeding a 500hz square wave, but the scope seems to be having difficulty in triggering.

[Martin.M]

that looks amazing, the PSU is 100% 

greetings
Martin

[Martin.M]

in that video. You have MISSING HORIZONTAL DEFLECTION LEFT HAND. 

so, see attached picture    or is the beam now allready working?

check the 6DJ8 please and also the 6197 there !

then give answer what you have found.

greetings
Martin

[Inverted18650]

I realize that this post if old but recently bought a 547 and I have not amplitude adjustments. I am going though this post line by line and doing the same testing / probing that was previously recommended. I will post the results if you all would be willing to help me troubleshoot mine as well. Thank you. Chris.

[tautech]

I realize that this post if old but recently bought a 547 and I have not amplitude adjustments. I am going though this post line by line and doing the same testing / probing that was previously recommended. I will post the results if you all would be willing to help me troubleshoot mine as well. Thank you. Chris.

Martin.M has a few of these old Teks, check out his thread:
https://www.eevblog.com/forum/testgear/vintage-tek-restoration-pictures-by-martin/

[Zalgar]

Hi mate,

If you had a similar issue to mine it was due to the -150V rail being off by about 2v. The adjustment for that rail is on the righthand side (viewed from the front) and down the bottom between two valves from memory. I no longer have the scope and have sold it to a bloke who uses it daily.

I would also check the valve and switch which connects whichever plugin module, as that commonly stops the ability to adjust the waveform from that module. I believe that was mentioned in this thread too.
The usual suspects aswell check that all the valves are glowing, and that the timed relay is clicking over its around 20seconds or so.

Good luck, I'd like to hear more.

Cheers!

[Inverted18650]

Hi mate,

If you had a similar issue to mine it was due to the -150V rail being off by about 2v.

Greeting and thank for the detailed reply and info. I gave myself a bit of a heat exhaustion injury (not my first) working on an outdoor project in this 95 degree weather and I'm stuck in bed resting for a few days. Once I'm feeling normal, I am start this restore. I also found a page by M.Martin (suggested by another user) which was incredibly detailed. I am super excited to tear it down and clean it, then build it back up to near mint condition. It will be my summer project. I'll take plenty of pics and even plan to do a few videos.
Thank you again,

Chris

[Inverted18650]

Got motivated and got up to take a few pics. The only tube not on is the one nestled in the back (K600- 6N030T) which I think I read was not supposed to be on expect under certain conditions - I have to read it again. I'll check the - 150V points and report back.

[Inverted18650]

Results:

All voltage were within tolerance and the -150 was spot on.

I am going to stop here and take Martin M's advice and clean everything before going any further. Once its all clean, I'll re test the voltages and report back. I am also going to print the manual and go through it again to make sure its all fresh in my mind.

[tautech]

Results:

-150V is spot on
+100V is low
+225V is spot on
+350V is low

I am going to stop here and take Martin M's advice and clean everything before going any further. Once its all clean, I'll re test the voltages and report back. I am also going to print the manual and go through it again to make sure its all fresh in my mind.

Sometimes I only print the bits you need, schematics and adjustment procedures are normally enough.
Voltages don't seen too low to stop it operating albeit it might be a bit out of Cal.
Good progress thus far. 

[Inverted18650]

I just re-read the manual and learned that all the voltages were actually within tolerance, so I updated the previous comment. ( I did so before I noticed you replied here, my apologies).

I also downloaded and started reading the Chuck Miller notes, which are incredible. I am excited for this project and hope I can get it working a dream again.