Hi group,
Next in line : the repair and restoration of my Tek 317.
Background:
------------------
That's my first tube based piece of equipment, probably the last as well since I don't intend to collect them, but I wanted to have at least one of them...
I was born in 1977 so not exactly of the tube era, so I wanted to have at least one instrument of this era so I can see it up close and fiddle with it, study it, and add to my little museum/collection of old stuff.
Electronics is basically 50 years of tubes followed by 50 years of solid state, so I can't possibly pretend to be interested i electronics, and wipe my butt with 50% of its history ! ... and I just love old stuff and learning about the history of science and technology/engineering, fascinates me, the ongoing march of the human race...
I have a fetish for CRT oscilloscopes and like Tek Scopes particularly. I have one for the early '90s, fully digital (TDS 544A), then a hybrid/combo one from the '80s (Tek 2232), and one from the 70's (Tek 5111) that's transistorized, so I wanted one from the '60s, tube stuff, to complete the collection. I wanted one that looks really old, with a round CRT not a rectangular one. So that would be early 60's not late 60's. The 500 series scopes seem to be by far the most popular, but they are very big and heavy as well as expensive, and I don't have much money to put in this purchase, nor much space for it in my tiny lab.
So, when I saw a local ad for a Tek 317, which I didn't even know existed, it was love at first sight ! It was small and beautiful to me, found it so cute ! Compact and light enough that I could actually consider keeping it in the lab to actually use it/play with at times, rather than use a hoist to put it in storage and leave it there sitting, out of sight, gathering dust. It looked in great nick overall, and even came with a few original accessories, which added value to it, in my eye. And it was cheap, 50 Euros... though it cost me a fair bit more than that to pay someone to pick it up locally (900kms away from me) and deliver it to me safely. But still, for a piece of Tek history, and a great piece of engineering, I didn't mind. Other thing going for it, was that the picture in the ad showed a trace on the screen ! So at least I knew it was not a basket case, which was good because I know zero about tube stuff, so I would be hard pressed to fix it !
I called the seller, said it was powered up recently when the pictures were taken. So, looking good indeed, so bought it !
Got it a couple months ago, now having time to work on it. In the meantime I learned some stuff about tubes, mostly thanks to that blog "Mr Carlson's Lab", a guy in love with anything featuring tubes, and very knowledgeable. Learned an awful lot of practical stuff, by watching his videos.
Actually his very latest video is particularly educational :
On YT there are also a few cool educational videos about electronics, that the US army did in the 40's, teaching the basics. Tektronix did something similar too, in the '60s. Came across one about tubes, which I liked, from 1943 :
Short but informative, and of course it's interesting to see how tubes were taught to people back when it was the current/state of the art of electronics, rather than a modern point of view.
The page on Wikipedia is interesting too.
Anyway, getting this old scope was also an excuse to try and learn about how tube equipment worked. I hardly intend to become an expert on the subject, but I wanted to grab at least the basics, so that they are not these magical and mysterious things anymore. I wanted to be able to make some sense of the schematics, follow the signal path, understand the structure of the various circuits.
The above material made that easy.. you gotta love the Internet.
My understanding
-------------------------
Basically it's quite simple. The tube behaves very much like a transistor, a FET transistor more accurately. With 3 active terminals (not counting the heater filament) anode, cathode and a control grid. There is current flowing from the cathode (which emits electrons thanks to the heater energizing it) to the anode, positively charged, which of course attracts the negatively charged electrons.
Just like a FET would control the flow of electrons in its channel by applying an electrical field between gate and source, that pinches the channel, the tube can control the flow of electrons by applying a negative voltage between the control grid (equivalent of the Gate of the FET) and the cathode (the "Source" of the FET). The drain is of course analog for the Anode of the tube.
So that's it. When looking at the schematics you just have to picture a FET and you suddenly find yourself being able to make some sense out of the schematics, magic... and you realize that it's very much like the modern electronics you are familiar with, same building blocks/basic structures, except that it was implemented using tubes rather than transistors, but that's about it.
What can cause confusion is also that some tubes have more than 3 active elements/pins in them... the basic type, the triode, has 3 terminals as its name implies, but on the schematics there are also Pentodes which as the name suggests, have 5 elements not 3. But nothing to worry about : fundamentally it works just like a triode/FET, it's just that it comes with 2 more grids which add refinement to it, but don't change the way the tube works. In practice, to understand the schematics, you can basically make abstraction of these 2 extra grids and whatever circuitry that surrounds them. The control grid we are interested in, is always represented/drawn as the very first grid (grids are represented as doted lines), counting from the cathode, which is usually at the bottom of the tube symbol.
So to follow the signal path, just look for the cathode, then the first grid from there is the control grid/gate of the FET, where the signal is applied.
The two other grids are there, as I understand, to improve the noise and bandwidth performance of the tube. The middle grid, second from the cathode, is called the "screen" grid. It helps reduce the cathode-to-anode parasitic capacitance, therefore improving the bandwidth of the tube.
Then the third grid, next to the anode, "plate" as it's also called, is the "suppressor" grid. Its purpose is to reduce the noise a tube intrinsically/by its very nature, generates, as the flow of electrons "crashes" and "bounces" off of the anode/plate. This is called "secondary emissions" (as opposed to the main/primary emission, from the cathode). The suppressor grid applies an electric field whose polarity and intensity, relative to the anode, tends to keep these naughty rebel electrons from flying too far away from the anode.
In practice these two grids are most often tied directly to the anode or cathode, or sometimes to an intermediate DC level using a simple resistor divider, and don't actually interact with the main circuitry/signal path. Very rarely (in this scope I mean, no idea about other types of equipments ! ), are those grids involved in the actual operation of the circuit. One example of that, (thanks to the great service manual which is very educational !) is the LV power supply section, where the screen grid is used to "inject" a signal to the tube, to modulate the input signal in a clever way, to cancel out the ripple voltage from the bridge rectifier.
So basically you have a simple triode/FET for most of the work, and when you require better performance on the noise and/or BW front, you use the extra refinement that the Pentode offers. For example in the preamp section, the input signal hits a pentode, presumably because like with any preamp you are worried about the noise polluting your fragile weak signal. Then once the signal is cleanly amplified, the output of the pentode feeds a simple triode, which acts as a driver/buffer to safely drive the next stage : the vertical amplifier section.
Sometime, maybe to save space, you would have the Pentode and Triode packed into a single tube assembly. A single tube can also host two identical triodes, which is handy when implementing for example a differential amplifier, also seen in several places in this scope.
So that's my basic understanding of these old tubes. I hope some tube expert will stumble upon this thread and correct me wherever needed. But so far this simple understanding of things proved efficient in helping me understand this old scope !
And I am glad to understand it because... yep you guessed it : it's NOT working ! LOL Yes, seller said it was working, and the picture showed a (flat) trace on the screen but.... I powered the scope and well, it kinda works yes, but no quite !
So I am glad in some perverse kind of way, because I will have now an excuse to really dig deep into the thing and spend time studying the schematics in detail, and probing all around the place ! Should be educational, and probably useful should I ever have to work on some other tube gear, though again it's not my intention to buy more of these.
So that's it for this long winded preamble/introduction. This old Tek is a work of art, the looks of it, inside and out, the build quality, the nicely laid out internals, the front panel that's elegant and beautiful, those lovely big knobs that feel so nice. This is a master piece of engineering and industrial design. I am so proud to own one of these, and I just can't get enough looking at it and playing with the knobs !
These old Tek scopes are a pure delight
So, needless to say I will try to fix it and restore/refurbish the thing.
Anyway, let's get to the meaty stuff now :
1) Repair : what's wrong with it and what have I done so far ?
2) Refurbishing : what's needed, what's planned ?
3) Some pics of course !
REPAIR
------------
Before I spend time and money refurbishing the thing, I first want to see if it can be fixed at all ! My priority is to have it working so I can play with it, then only do I want to make it look good.
So, the symptoms are as follows : You power up the scope, it appears to work fine (tried all basic controls), but instantly, the trace on the screen starts to move up and up and up... and eventually disappears off the screen. At the same time, the brightness of the trace progressively decreases, so much as to eventually become completely invisible. Cranking up the brightness/intensity control only buys you some time, but not much at all. It all happens quite fast : in practice, once you have powered up the scope from cold, you don't get much time to play with the controls, before the trace is gone. I will try to make a short video to better show what I am talking about.
From what I can see, it can display a sine wave just fine, clear and crisp trace (focus and intensity work fine), triggers fine, vertical and horizontal position controls work, time base works, It basically works just fine... if only the trace didn't feel the urge to run away and turn to black ! LOL
My take on it : the fact that the scope works fine when cold, kinda leads me to believe that all the tubes are fine. I reseated all of them for good measure anyway, and not surprisingly it didn't change a thing. From what I understand, the failure mode of tubes is mainly a progressive degradation of performance due to imperfect sealing of the tube (ambient air being sucked in, seeping through the pin holes at the bottom, the glass not making a perfect bond with the terminals). But whatever problem this might cause, I think it would happen from cold, right from the beginning, and it would be constant/stable. This is not the case here : scope works just fine when cold, and the problem changes with time.
A trace drifting could be a warm up issue... but I don't think this is it, because it happens way too quickly ! So I think it must be due to some capacitive effect. A bad capacitor somewhere, or a failed resistor that makes a capacitor not work not in the way it was supposed to. But something capacitor related, I think. Because the problem makes the trace drift vertically, I assume the problem must lie in the vertical amplifier or preamp. Either that, or a system-wide problem, ie the power supply. Had a first look at this side of things. IIRC there are 5 big "cans" in the thing, each of them actually hosting several caps. I see that people are used to "restuffing" these cans with new, modern low ESR electrolytic caps. I guess it would be good measure regardless. I don't have an ESR meter just yet, but could try and measure their value to get a clue maybe. Don't have a leakage tester either. Whatever, if the filter caps go bad so much that it makes the scope go nuts as much as it does, then surely it should affect the power rails greatly and this should be easy to detect just by measuring them and checking for ripple.
I gather that very old tube gear had paper&wax caps in them, that invariably leak and can cause all sorts of very serious problems because they don't block DC anymore, but this scope is not "that" old, it hit the market in 1959, and it being a Tek I guess they would have used the best caps available at the time. I would need to give it a closer look, but at first I didn't see any of these caps in this instrument. So, if there is a failed cap(s) in this scope, I don't think it would cause a catastrophic failure. I am not too worried about that.
Then there is the problem with the trace intensity getting gradually dimmer and dimmer. Again, the progressive nature of the fault points to a capacitive issue, in the HV/CRT section this time.
I have ordered an HV probe so I can poke around that area, but while I am awaiting delivery, I can start probing all the LV rails with my DMM. They range from -150V up to +420V, which a DMM should be able to manage/measure safely, I would think.
Service manual : too big to attach here, 60MB in all. tried to pull the schematics out of it but that 'still too big to attach (4MB or so). I might attach single/individual schematic pages though, as I go over the various sections one by one. Manual can be downloaded on this page, from a Tek related website. The page also gives some info about that scope :
http://w140.com/tekwiki/wiki/317Direct link to the PDF :
http://w140.com/mmm/tek-317.pdfRESTORATION
-----------------------
Overall it's in great nick, no major issues. I mean, the case is not damaged, and no parts are missing, and there is no corrosion inside, and only a very minimal amount of dust inside. Would still like to clean its guts though, to make it all nice and shiny.
At the back, the air filter has seen better days. It looks dark yellow, like it was filled with honey or something... brrr....
Should be an easy fix though : the manual says that the filter is made of a metallic "wool", and is designed to be washable. So, let's wash it then...
At the front, looking good too. The front panel is in great condition, the lettering is in top shape, no buttons are missing nor damaged. It just needs a good clean. There is however a large/long scratch at the bottom left part of the panel, in the vertical section, but it is very light/superficial, not really a show stopper. It could probably buff out just fine but I fear this might cause slight change in the surface finish, which would be noticeable, and overall do more "damage" than anything else. So for now, I think I will just leave this scratch as it is...
There are a few bits and bobs that need some attention too :
The 3 coax connectors, which are "UHF" type rather than BNC, are rather crusty and one of them is even missing its dielectric piece ! So I ordered new ones. Looks however like changing them won't be fun, access to two of them looks somewhat challenging so to speak...
Most UHF connectors I see for sale, have a square shaped/4 hole base plate, but the connectors in the Tek are 2 holes things. 4 hole connectors would not fit, it would not clear the surrounding metal work. Luckily I came across on Ebay, some 2 hole connectors, which therefore should be a direct fit, I am hoping. But they are cheap stuff from china, so I have yet to receive them but when I do, if I am not happy with the quality of them, I would have no choice but to buy quality stuff from known manufacturers like Amphenol or TE connectivity, but they only come in 4 hole versions. So I would have to saw/cut two of the corners to trim them so they can fit the chassis.. but no big deal, and actually, looking closely, it seems to me that it's exactly what Tek did back then ! So if they did it, why couldn't I...
The UHF-BNC adapter (required to hook the probe), is starting to show its age, nearly 60 years now. They are still made these days, so I ordered one of those too.
Then the neon indicator bulbs (for the UNCAL positions of the knobs) look tired to me, though I have no experience of them to be honest so can't really say for sure. Still, these appear to be standard items, called "NE2" bulbs, which are still available today for little money, so I bought a few of them, will see if that makes any noticeable difference in brightness terms.
Then last : the cooling fan. it doesn't start right away. When you power up the scope, it takes a minute or two before the fan decides it wants to start spinning. When it does, it is very quiet and with no parasitic noises. It's a silk smooth ride, quite impressed I must say ! Quite worried too : it's "too" quiet I think... I would expect such an old fan, in such a power hungry instrument, to be quite noisy. So I fear that it might be "tired", somehow.
I checked the schematics just in case there somehow was a timing mechanism somewhere, for some reason that escapes me, to purposefully delay the spinning of the fan. But no, no such thing : the fan is connected directly across the primary windings of the transformer, so it ought to start right away.
Well, it's not STRICTLY connected directly to the transformer... the schematics show a resistor in series with the fan. So maybe that has drifted over the many decades, and gone up in value, restricting current flow, which would explain things. Will check for that.
I think that's about it for the initial rambling/introduction. Sorry to all those who fell asleep before reaching the end,
and congratulation to all the others, you deserve a pat in the back
I will now start working on the thing and post as I go, as always. Hopefully some Tube Guru will come by and offer guidance whenever I am stuck, out of ignorance. I do have a clear path of investigation though, to get me started, and plenty of things that I can check, so hoping to make some progress by myself and bother the tube gurus only when I really run out of ideas.
PICS
--------
I opened the cabinet to expose its guts, from all 4 sides. As for the original accessories, you can see the probe and the inside of its body (a nice die cast aluminium construction !), the UHF-BNC adapter that goes with it, and.... and.... this huge eye piece, no idea how it's called exactly, which I think is used to see dim traces when the ambient light is strong. Especially useful I guess, because this scope as I understand it is designed as a "portable" unit to be carried around and used outside in the field, rather than in a comfortable and well controlled lab environment, which is more where the more popular 500 series scopes would be used, typically.... I think !