Tektronix 545A troubleshooting

[rivet_head]

I was given a 545A some time ago and finally decided it was time to get this boatanchor up and running.
I got her running through a variac and isolation transformer @ 117v AC, fan runs, power light is lit, looks like the heaters are all glowing.
The start delay relay (K601) is not clicking over, I'm also only showing 5.5v on the -150v test point.
Any insight as to where to go check next?

[rivet_head]

http://bama.edebris.com/download/tek/545a/tek-545a.pdf
page 123


All diodes in rectifier stages have approx 0.5v voltage drop, none are shorted. 2 resistors coming off one of the rectifier stages look a bit iffy, will be replaced.
Cap 601 shows 48Ohm on esr meter, capacitance not measurable.  Tek parts lists says 0.1uf (100nf) 500v "discap".   Forgive my ignorance, but what the heck is discap? discap or that cap? discharge cap?

[Gene]

Hi,
I had a Tek 535 that behaved in the same way: switch on, sound of the fan, no clicking sound of the relay.
It turned out that the coil of the relay K601 was not energized because the thermal delay relay K600 was not properly inserted in his socket.
The -150V rails shows up only after K601-2 is closed (the switch near the full bridge rectifier D642 ABCD) so I would not be worried too much.
Cheers,
              Gene

[Dave Wise]

http://bama.edebris.com/download/tek/545a/tek-545a.pdf
page 123


All diodes in rectifier stages have approx 0.5v voltage drop, none are shorted. 2 resistors coming off one of the rectifier stages look a bit iffy, will be replaced.
Cap 601 shows 48Ohm on esr meter, capacitance not measurable.  Tek parts lists says 0.1uf (100nf) 500v "discap".   Forgive my ignorance, but what the heck is discap? discap or that cap? discharge cap?

C601 is a ceramic capacitor in the "disc" form factor.  Somebody thought discap was a clever portmanteau.

0.1uF is small compared to the caps your ESR meter was designed to test; unless the cap is bad which is quite unlikely, that 48 ohms is mostly reactance, not resistance.

The time delay relay circuit prevents the power supply rails from going live until the tubes have had time to warm up.  If the warmup is bypassed or not all rails come up at the same time, some tubes will see cathode-to-heater voltages far above their maximum ratings and may flash over.

Tek chose some really fine parts.  To antique radio restorers' astonishment, the electrolytic caps are usually okay, although the first power-on may draw excessive current due to the caps having lost some of their oxide insulation layer after years of disuse.  Don't freak out if the mains fuse blows the first time.  If you want to re-form the caps on an external power supply, disconnect them or remove all the tubes.  If you remove tubes, label them to put back in their original sockets so as not to make re-calibration necessary.

[rivet_head]

Thanks Gene - I pulled K600 for cleaning and eyeball inspection during my initial once over cleaning. I'll pull it again and take a closer look at the socket, pins, and insure shes seated properly. I pre-emptively ordered a new 6NO45T and K601 relay a few days ago...found 'em for a very reasonable price on fleabay.


Dave - DISC CAP!  Holy crap I feel dumb. I appreciate the candor.      I figured the 0.1uf was below the ESR70s measurable range. Just checked the book and yeah, minimum is 0.3uf. I have an EICO 950b I recapped that I believe can measure as low as 10pf (book says 10mmf), but I'll have to pull it to test on the 950.   I've pulled most of the tubes at least once in the last month or so for cleaning and inspection. The chassis was THILTHY with 'ive sat out in an old shed for 25 years' dirt and dust.   When I first got the unit (10 years or so ago) it powered up and I remember the distinctive CLICK of the relay and the rest of the unit coming to life. The CRT image was a smudge/blur of a trace that "paused" in the middle of the CRT. At the time I felt I lacked the necessary skill to correctly troubleshoot her.

Here's a question. Do K601-1, K601-2, K601-3, and K601-4 represent the 4 poles of the relay?   
AND
In the parts list k601 is listed, for my serial # (36000 something), as DC relay coil, 2000Q, but right above it serial #s 20001-27729 list k601 in the usual 4pDt 6.3v fashion.
This confuses me.

[Gene]

Hi,
I think that all the K601-x switch are ganged together. The rationale (as my father told me several decades ago): "If the high voltage is applied to the the anode before the cathode is hot enough to provide enough electrons, you can easily damage the tube".
The K600 switch closes when its heater is hot enough. This should guarantee that all the tubes are operational so K601 is energized and it closes K601-2 etc.
If anything goes wrong K601 is de-energized and a second attempt occurs after 15 seconds or so.
You can find here:
https://www.ebay.com/itm/234354274044?nma=true&si=hkJvSIUHNc2kVGAQvufg9q8fvRI%253D&orig_cvip=true&nordt=true&rt=nc&_trksid=p2047675.l2557
a picture of the relay  148-016 (the one you should have)

and here https://www.ebay.com/itm/334451743828
a picture of the relay 148-004 (the one mounted on older Tek 545A).

The pin layout seems a little bit different, perhaps you can have a look and identify which one is mounted on your scope.

All the best with your 545A, 65 pounds of pure electronic ecstasy,
Gene

[rivet_head]

Thanks for clarifying.
I pulled the cover off k601 whilst cleaning, so I atleast had a visual reference while hunting down the part. Glad I did.
https://www.ebay.com/itm/254190029500
Is what I should be getting in the mail any day now.
Your explanation of the 6NO45T makes perfect sense now that I'm looking at it.   The contacts look like they have a little carbon on them. And the (I'm assuming) heater connected to pins 1 and 6 looks to have a bit of discoloring. What do you think?

[rivet_head]

Heres the heater piece and the discoloration

[Gene]

Hi,
perhaps you can test your 6NO45T following the procedures described here:

I remember that my initial suspect was the relay K601, since it is quite difficult to replace it, I turned on the 545A by a gentle touch with a plastic rod on top of the relay armature.
Once you tap on the armature the coil of K601 remains latched by the action of K601-1 that at the same time disconnect the heather of K600 and bypass its contact.
Because of the high voltages (and of the high currents) please don't do that 

In the next days I will have a look at the 6N030T  (similar to your 6N045T) that is inside the Tek 547 I have in my office.

Cheers,

Gene

P.S. I really like the way you are referring to your 545A, yes she is a scope, the Lady of the scopes.

[jdragoset]

If it does not test:
https://www.ebay.com/itm/283449635099?hash=item41feea311b:g:QMYAAOSw-olaHcS-&amdata=enc%3AAQAHAAAAoBcZm0aFFdhPzG4AUcqiHTXDAEOn2Z9TI2XKUnFUaJQaniy2AVxTyzwwDgSbbl6Z%2Bd6kRvJZ%2FWBHr%2BkmnWLxPJizw3p9l7AQALZAItUjSMlTCoAXdIpPatwOb3XwQ5%2Bnfz0wlgipoFm%2Fhq3JYImngI1iAGcAZChKG1ft%2FVRaqHTaAcC%2FlU3TBUb7iC0%2FicVydOMA2IB%2F9LyAgfJmG%2BcJINA%3D%7Ctkp%3ABk9SR6C5yLLGYQ

[Dave Wise]

K601 originally had a 6.3VAC coil.  The mains-frequency magnetic field around the relay affected the trace.  They fixed this by switching to a DC-powered relay.  This is documented in the 535A Mod Summary (available by contacting VintageTek.org) as mod kit 040-0258-00.

I have a factory prototype 535/535A.  Its naked AC-powered relay was wiggling the trace.  Rather than do the DC mod, I just added a mu-metal box, gift of the late Stan Griffiths.  That made it good enough but Tek went the extra mile with DC.

Cathode stripping is a thing only in high-power tubes like in transmitters.  In the Tek scopes, the problem being avoided is over-voltage between heaters and cathodes.  Look at some of the long-tail pairs and you'll see what I mean.

[Gene]

Hi Dave,
thank you for the clarification on the differences among the two relays. I guess that the metal box in the auction is the mu metal shield you are referring to.
I see your point on the heather to cathode voltage rating, still I think that my father concern on operating the warm tubes is meaningful at least for the two 6080s V737 and V677 that provide the 350V and the 225 V rails.
Anyway it was more than thirty years ago when my father told me this. I was a teen ager and perhaps I misunderstood.
Still I wonder how the Tek guys solved the timing problem on the 556 which does not click after power up and, at least to the best of my knowledge, lack the thermal delay relay. As a mater of fact, the first time I powered it up, I was afraid of something bad till the traces showed up.
Thank you again for the clarifications and the for informations. All the best,
   
   Gene

[Dave Wise]

Gene, thank you for bringing the 556 to my attention.  I will study this and report back in a couple days.  The Lecture Notes mention that there's no time delay relay, but they neglect to explain why.     There are no 6080's, which supports your dad's theory.

[rivet_head]

So I tested the 6NO45T with an old National Tattoo power supply. 6.3vDC across the heater, continuity test across contact pins. Contacts slowly crept towards close, ~45 sec later contacts closed. cleaned up the pins and pin sockets for the 6NO45T, reseated, 117v supplied to mains, power sw on, and no change. I can see that the contacts of K600 close in the desired amount of time, but NOTHING on the K601 coil. Not a spark, jump, wiggle, nothing. Most, if not all, of the tubes had heater glow.
After about 2 minutes I took a long non-conducting force applicator and gently gave the top of K601 a little assistance closing, hoping she would spring to life.
I saw a very small spark/arc on each of the K601 contacts, noticed a few of the tubes in time base A, time base B, trigger A, and trigger B were now at "operating glow" but no change anywhere else.   Turned it off and tried the same thing with the plug in removed.   The only change was that one of the trace indicator lamps was lit up, saying the trace was off the crt. No front panel controls would change that lamp, and no sweep on the crt.

I just got done tweaking the contact arms on K601 and cleaning up the heads on the contacts, gotta run some errands then ill see if anything changes.


I noticed this a few days ago but never saw just how horrific it really was until now: the bottom of the trimmer c375 that hugs the Mag Gain pot (r375) does anyone have a picture of what the underside of c375 SHOULD look like?

[Gene]

Hi,
after the gentle touch on top of K601 did the relay latched in the closed position without further assistance? If yes the coil of K601 is working properly. Perhaps the top armature of the relay is oxidized and got stuck?
I would focus on the power supply before looking elsewhere for bad components.
Cheers,
Gene.

[rivet_head]

No K601 would not stay latched.

I did notice that the 4 lamps B434A B434A (in delay circuit) B347 and B386 were lit when she was first turned on, but immediately faded to off

Test point voltages @ 117vAC line, drawing 2.4Ah, no test unit installed. from front to back ( L to R) =
6.3 = 12.4v 6.3 = 12.4v    6.3 = Zero   -150 = +5.7v   +100 = +130v   +225 = +12.5v   +350 = -0.6v   +500 = +218v

[rivet_head]

well decided to go over the rectifier stages again and low and behold I went to desolder r700 due to some discoloring and  only half came out.   Replaced w 10 Ohm 2 watt resistor. No change on behaviour,, very miniscule change, if any, to voltages. The other 1Ohm resistors coming from the rectifiers only measure 6 Ohm. currently the only other 10 Ohm resistors I have in house are 10 watt ceramic big boys.

[Gene]

Hi Rivet Head,
I remember one of these 10 ohm resistors burning and producing the magic smoke on the Tek 535A I had  :'(.
If R700 was discolored and overcooked, it could be useful to have a look at C700 and to the other 10 Ohms carbon resistors in the power supply section.

In particular I would keep an eye on the rectifiers in D642, the resistors R640 and R641 (they could be as bad as the one in the picture you posted), and C640 of the -150V section because all the other voltages provided by the power supply are referred to the -150V rail. If the -150V rail is at +5.7V all the other rails will go southwest.

Cheers,
Gene

P.S. are the resistors measuring 6 ohms R640 and R641? These are in parallel, hence 6 ohm is perfectly fine.

[Dave Wise]

Cathode stripping is a thing only in high-power tubes like in transmitters.  In the Tek scopes, the problem being avoided is over-voltage between heaters and cathodes.  Look at some of the long-tail pairs and you'll see what I mean.

RETRACTION

I asked one of the designers of the 556 about this and here's what he said: "As far as I know, it was mostly the vacuum tube series pass regulators losing their oxide coatings. I've seen it and it's not pretty.  Physical craters appear on the oxide surfaces."

That said, there are places where heater insulation is stressed to or beyond the maximum rating: V1174A, V1184A, V1274A, V1284A, V534, V673B, V734, V873B, V1104, V1194, V1294, V605.

[rivet_head]

Hey Dave, Thank you!
I took a second look at the rectifier stages...    My first pass through them I only checked forward vdrop and continuity, and never reversed my leads across them.  Looks like both the +100 and -150 supply diodes are shot as I see ~1v across them after reversing my leads. I have some 1N5406 diodes on hand that are 600vr 3A (The 1N2862A diodes in the 545a are 400vr 1A I believe) so ill get to swapping those out in the next night or so. The rest of the resistors in the supply stages look, and measure alright. I gave em all a few pokes too, incase any others were "working" but split in half 
You guys might find this humorous. I took the clear plastic cover for the supply rectifier stages off a few weeks ago and set it aside. Since then I've used the manual and printouts for all my component referencing back and forth, back and forth. Well wouldn't ya know...the clear plastic cover has the supply lines, rectifier stages, and resistor names printed on it RIGHT OVERTOP WHERE THE COMPONENTS ARE. I swear I like to do things the hard way sometimes.

[rivet_head]

Update.    I tested C640 and got 93.2µF, ESR of 0.14Ohms. Caps stamped 125µF 350v.   Waiting on its replacement to get here.  I honestly forgot there were two caps sandwiched between the back frame and transformer. Had to remove the bracket the bracket with the 7 big wirewound resistors on it.

So maybe someone can enlighten me here. I tested C640s neighbor, C649, while I was at it. ESR meter gave me a value of 82.92µF. I checked the parts listing and its shown as 2 x 40µF, but the schematic for the power supply (same book) states C649 as 2 x 20µF. Is the schematic based off the original design and just missing an updated component value? There are components with values highlighted in red in the schematics to show that they have been up updated, but C649 is not one of them.
Thanks in advance.

[Dave Wise]

I believe that "2x20" is a schematic error.  The parts list shows the part number and value of 2x40, and the 535-545, 535A, and 545B manuals concur.
Since none of the manuals show 2x20 as an early value replaced by 2x40, it's probably a straight-up mistake.  I don't know if it was ever corrected; it's wrong in my manual for serial number 39010.

[rivet_head]

Sorry for the delay in update. So after replacing a few caps and resistors I've got -150v at the test point! the -150v adjustment pot works as well. All other voltage test points are spot on, or within 2%. Except for the HV test point. I have no trace on screen still, but the indicator lamps are acting like there is a trace. The indicator lamps also respond to horizontal and vertical changes. The HV circuit in this unit sill has the black beauty sprague caps, so I'll start with replacing those. The HV transformer and tube/valve "diodes"? light up my high voltage indicator screwdriver as soon as it gets within an inch or so from it, so I'm assuming that part of the HV circuit is working.

[Dave Wise]

Here are the notes I compiled for my own 545.

SUGGESTED PART NUMBERS
Nominal specs from 545
In stock at Mouser 25-Mar-2021

There is about 20mm vertical space between the baseplate and the tubes.
(About 15mm on the right side due to filament wires.)
Box film caps fit just fine provided the three lower-voltage
parts are skinny so you can stack them.

Grid/Feedback/Cathode: 6.8nF 3kV, use 10nF 3kV box film, Wima MKP1W021005B00JSSD $0.89 6x15x27mm
If you allow 2kV, you can use Kemet F463DB103K2K0Z, R76UN2100SE30J or R76UI182050H4J,
or Epcos/TDK B32672L8682 or B32672L8822, or Nichicon PPB2201680KGL.

Tripler 1: 6.8nF 5kV, use 10nF 6kV box film, Wima FKP1Y021006F00KYSD $1.83 15x26x32mm
Or 6.8nF 6kV Wima FKP1Y016806D00JSSD
Or 10nF 4kV Wima FKP1X021006F00KSSD
Or 6.8nV 4kV Wima FKP1X016806D00JJ00
Or 10nF 6kV Z5U ceramic, Vishay 564R60GAS10, $5.86 ouch

Tripler 2/3: 470pF 10kV, use 1nF 15kV Z5U ceramic, Vishay/Roederstein HVCC153Y6P102MEAX $2.44 8x15mm
Or 1nF 7.5kV Z5U, Vishay 564R75GAD10 $1.90
Or take your chances with Chinese sellers at eBay.  I've
seen great parts and crap, and there are no specs.

PLACEMENT/ASSEMBLY
Between the two rows of ceramic strips except as noted.

On the left, place Grid flat with leads facing left.
On top of Grid, place Feedback with leads facing right.
Place the feedback resistor string to the right of Feedback.
On top of Feedback, place Cathode flat with leads facing right.
Near the upper right, place T1 flat with leads facing left.
Above the upper right strips, hang T3 vertically.
Below the lower right strips, hang T2 at an angle, head down/leads up,
avoiding the ground lug and straddling the filament wires.

Grid/Feedback/Cathode caps that are short enough to lie on their
backs can be installed that way.  The baseplate is 70mm wide.
T1 is 26mm + 3mm for wire bends.  The feedback resistor
string is 5mm wide.  That leaves 12mm per cap, which opens
up more part numbers, especially if you include 2kV.
We're still single-sourced on the tripler.

Grid/Feedback/Cathode:
3kV Wima      MKP1W021005B00JSSD $.89  6x15x27mm
    Panasonic ECW-HC3F822
              ECW-HC3F103        $1    8x16x27mm
2.5kV Kemet   F464BP103J2K5A     $1    9x15x18mm
     
2kV Wima      MKP1U021004D00KSSD $.70  7x14x18mm
              MKS4U021004D00KSSD $.68  7x14x18mm
              FKP1U021006B00MSC9 $1.45 11x21x32mm
    Epcos/TDK B32672L8682J       $.67  6x11x18mm
              B32672L8822K000    $.70  6x12x18mm
              B32613A2103J008    $.84  11x17x27mm
              B32652A2682K000    $.94  11x19x18mm
              B32653A2682K000    $1.30 9x17x27mm
    Kemet     R76UI210050H7J     $.76  12x13x18mm
              R76UI182050H4J     $.67  6x12x18mm
              R76UN2100SE30J     $1    6x15x27
              PHE450SB5100JR06   $1    7x13x18mm
              F463DB103K2K0Z     $1    6x15x26mm
    Nichicon  PPB2201680KGL      $1.30 6x15x27mm

On the left, place Grid and Feedback on their backs.
Place the feedback resistor string to the right of Feedback.
To the right of that, place Cathode on its back.
Near the upper right, place T1 flat with leads facing left.
Above the upper right strips, hang T3 vertically.
Below the lower right strips, hang T2 at an angle, head down/leads up,
avoiding the ground lug and straddling the filament wires.

IN THE SECONDARY HV COMPARTMENT

Original was 15nF 3kV.

3kV:
Wima FKP1X021507C00MSSD $2.32
CDE 940C30S15K-F $3.36 axial

2.5kV:
Kemet F464BS153 $1.47

2kV:
Kemet PHE426SB5150JR06 $.83
Kemet R76UI2150SE40J $1
Epcos/TDK B32672L8153 $.99
Wima MKP1U021505B00KSSD $.85

I'd just glue it to the chassis and run wires to the strips.

[rivet_head]

Dave, I appreciate the info, would you mind explaining in a bit more detail? how many of each value did you need to get, and for what C# on the schematic? I see a small mess of black beauties underneath the 5642s. I think there are 5 (?) caps under them to be replaced.