No horizontal scan on Tek 465B

[Radio Wrangler]

No, you've still missed something. Check the emitter voltage of each transistor with each of the timebases selected.

Q4465 emitter is 90mV when timebase A is selected. This is good. It shows that Q4482 is forcing Q4465 off, which is what we need at the moment, but checking Q4465 emitter with timebase B selected is needed to see that the B timebase switch is working normally, and that the off-looking voltages aren't due to failed parts.

Q4475 emitter at 0.7v looks like it's working and turned on, and the measures +21v at test point 72 accounts for your strong left shift. The collector voltage isn't very important just now, but we need to check it in the off condition, press B and measure Q4475 emitter, it should be below 200mV.

You've verified that one switch is off and the other one is on, but you haven't measured the opposite condition.

What I'm now wondering is why there is a strong left offset in XY mode. Does the scope turn both A and B off when XY is selected, or does it park one timebase at +13v and allow it through to the X amplifier, and have the earlier stages provide a countering offset?

I think it's the latter. U4491A has only A on or B on states so only one timebase can be connected to X at once, never both never neither.

This means that one timebase, probably A, or perhaps it doesn't matter which, is selected, but the sweep generators are parked at their waiting voltages, +13v.

And now seeing +21v on test point 21 starts to explain the left offset in XY mode. more positive means more left.

So the next question is why it isn't holding at +13v?

This means looking at the feedback loop Q4588A and B, Q4497, Q4498, QQ4598.

Is Q4588A base at +13v? If the pot R4570 slider has gone open, Q4588A would be stuck off, test point 71 would TRY to go down low, Q4497 would clamp it still at about -4.7v but by passing extra current through Q4497 which would turn Q4498 off, and Q4598 off, letting Q4598 collector go very high.

Possibly there are dead parts in the loop , or maybe something in the A SWP start gate circuit off of this page is hauling current out of Q4497 emitter.

You should be getting the idea that a lot of this circuitry works by signal currents more than voltages, so we need to look at places where voltages give away what the currents are doing.

David

[465-owner]

Good afternoon David,

Thank you for your early start and for your additional comments and suggestions.
These are some of the measurements I think may assist.
Q number,[Timebase A] , Emitter, Base , Collector , [Tb B], Emitter , Base , Collector,
Q4465 , [A] , 90 mV , Ground , 120 mV , [B] , 700 mV , Ground , 490 mV
Q4475 , [A] , 730 mV , Ground , 120 mV , [B] , 60 mV , Ground , 490 mV
Q4482 , [A] , Ground , 760 mV , 88 mV , [B] , Ground , 144 mV , 730 mV
Q4481 , [A] , Ground , 160 mV , 730 mV , [B] , Ground , 750 mV , 60 mV
U4491A pin 9 , [A] , 3.4 volts , [B] , 230 mV ,
U4491A pin 8 , [A] , 3.3 volts , [B] , 3.5 volts, *
* There seemed to be a problem here then I discovered it goes low (220 mV) with 'Intensified' or 'Alternate' buttons pushed.
U4491B pin 5 , [A] , 160 mV , [B] , 3.6 volts
U4491B pin 6 , [A] , 150 mV , [B] , 3.6 volts

I don't know if it is relevant but Test point 72, Collector 4598 is 21.4 volts with 'A' timebase selected and drops to 20.6 volts with 'B' selected. There is no change in X-Y mode and the voltages are still 21.4 and 20.6 with A & B respectively.

Q number , [A] , Emitter , Base , Collector , [B], Emitter , Base , Collector,
Q4588A , [A] , 13.6 V , 12.9 V , Ground , [B] , 13.6 V , 12.9 , -290mV , *changes to -3.5 in XY mode*
Q4588B , [A] , 14.4 , 21.5 V , Ground , [B] , 14.4 V , 20.6 V , Ground

Q number , Emitter , Base , Collector,
Q4497 , -290 mV , -980.mV , 15.6 V,
Q4498 , 790mV Source? , 15.7v gate, 15 V Drain?
Q4598 , Ground , 620 mV , 21.5 V ,
Sorry I wasn't sure of the FET's drain and source.
All above measurements except where indicated are with A timebase only at 2 mS / Division.

Again, thank you so much for all the time you've spent on this for me.

All Best Wishes

Nigel

[Radio Wrangler]

Let's look at Q 4588a it's base is +12.9v which is good.
Q4588b base is +21.5v which follows the voltage on TP72

Q4588B should be completely shut off. Now Q4588A emitter is at +13.6v which is right. The two diodes in the emitters are to stop the base emitter junctions being destroyed by reverse voltage. Typically they're rated at 5v reverse, usually they undergo zener breakdown around 7v. Single easiest way to destroy a transistor.
If you put a voltmeter on Q4588B, I'd expect CR4591? the right hand diode to conduct and for you to see about 13.6v, except Q4588B base emitter would be zenering from +21.5v on its base and giving you about 14v.

I think replacing Q4588 A and B along with the two diodes and checking those two 36 0hm resistors.

Just a pair of general purpose 30v PNPs will do for testing in place of the dual transistor. 2N3906 or something like that. 1N914 general purp diodes will be perfect.

Q4497 emitter at -290mV with Q4588A carrying all the current is reasonable. The diode across Q4497 is there to protect against reversed base-emitter voltaage. The base voltage of Q4497 should be -4v from those two resistors, but that protection diode is doing its job and is pulling the base bias upwards.

Q4497 should be thoroughly off and its collector has risen to Vt. Seems OK.

Q4498 isn't working as a JFET, with that large gate voltage, the gate to channel junction diode is ON, and is clamping the gate to the +15v rail, with an extra 0.7v for the diode drop.

THe FET can't pull its source above 0.7v because of the base-emitter diode of Q4598

So a gate to source voltage of 15v on the JFET looks like it's kaput. Q4598 with 07v on the base and still tit's collector up at +21v looks dodgy too.

I think there's been a failure like a stack of dominoes.

You may want to replace that whole loop of transistors as a oner before applying power again.

David

[465-owner]

Thank you David,

I only have a very poor selection of silicon PNP transistors but found five BC178s, two of which I used to replace Q4588 A & B but without any change in voltages. I checked the two 36 ohm resistors and also confirmed that there is 0.6 volts (from DVM) across each of the two diodes with transistors out and scope power off.
I also temporarily replaced Q4497 with a 2N3704 but the same voltages were around it too.
The only FET I have is an N channel MOSFET VN10LM but I don't know if I would do more harm than good by trying it. Regrettably I don't know anything about FETs.
If one or more of the transistors were dead wouldn't one expect some of the voltages to be different when a new one was put in?
The part number of Q4498 in the manual equates to SFB8129 which is a T.I device but it can't be found with a search of ti.com. Below is a copy of the relevant row of the manual:
A4Q4498 151-1025-00 Transistor:Silicon,JFE,N-Channel 01295 SFB8219
The VN10LM (RS stock No 295-107) is brand new in an RS bag but the RS web site can't find it!

Thank you again for all your help and suggestions.
Best Wishes
Nigel

[Radio Wrangler]

Just in and shattered, been out curling tonight (first time ever) but it's been a 70 mile drive home.

The circuit is a loop, and one of those things where a fault will destroy other semiconductors, so replacing transistors singly can become an endless task.

VN10 is an enhancement MOSFET, N-channel. It is normally off and needs positive gate-source voltage to turn it on. The FET needed is an N-channel J-FET. It is normally on and needs negative gate-souurce volts to start turning its current down. They are very different beasts.

David

[465-owner]

Thank you again for your time and trouble David.

I regret my time has had to be given to more mundane but more essential tasks and I haven't had time to 'play' with the scope any more. I doubt that I will be able to again until next week.
I spent a few minutes looking for other FETs, I thought I had a used 2N3819 somewhere but couldn't find it. If I do find it should I try in the position of Q4498? Would it be compatible?

I wondered if it would be useful to try an external sawtooth (approx 4v p-p?) applied to the the base of Q4598 via a 10nF Capacitor?

It's nearly weekend again I assume you will be going back to your roof repairs.
Thanks again for all your help.
All Best Wishes, Enjoy your w/e.
Nigel

[465-owner]

Greetings again,
After a long time with no progress because it took ages to find and obtain the CA 3046 and the 2N3819s. Both RS and Farnell have discontinued supplying them. I eventually found what I think are Chinese 2N3819 J-FETs from an Ebay source in England. I had asked several television dealers and repair shops without success. One dealer repair man had never heard of an N channel J-FET and asked what it was!
I'm pleased to report that changing the following has made an improvement although not solved all the troubles.
U4269 --> 2 X BC178
Q4588 --> CA 3046
Q4497 --> 2N3704
Q4498 --> 2N3819
Q4598 --> 2N3704
Q4475 --> 2N3704
The trace spot can now be centred and adjusted horizontally with the position control. Centering it gives 46 volts on both of the deflection plates.

I still have no proper horizontal deflection although under some conditions there is a very small and non-linear deflection which is unaffected by the time per division control. It does give nearly 3 cm deflection with with the times ten multiplier pushed and is affected by the trigger hold-off control. I suspect it might be some sort of instability, It seems to only show up for the first few minutes after the 'scope is turned on. It only occurs in Auto sweep mode, not in Normal mode. Waveform investigation at Connector J4571 pin 1 shows the instability to be there when it exists so I would assume all the following circuits are now cured.
The DC voltages around Q4497, Q4498, Q4598, Q4475 and the pair Q4588 are all showing as indicated in the Tek 465B circuit diagram.

I then went back to Q7254 & Q7256 which seemed to have a trigger pulse but it must have been related to the instability or whatever as the timing does not change with the 'A' time per division control (only with Trigger Hold Off).

I trust the photos will help, the last one (1197) is at a very odd angle but the papers are horizontal.
I note this evening the 'instability' deflection has remained on on screen for an hour although it isn't any cooler or warmer than earlier.

I will look forward to comments and suggestions or perhaps suggestions as where to go from here.

[465-owner]

Greetings again,
After nearly a month I have at last been able to return to my 465B. I must first apologise for the error in the list of component changes:
U4269 --> 2 X BC178 and Q4588 --> CA 3046 should be the other way round. U4269 being the CA 3046 and Q4588 being the pair of BC178s.

Since then I have reverted all the changed components one at a time, checking each time and the only component which makes any difference is the Q4588. It is possible that I may have killed it as there was no deflection or even dot. Display was as it would be if TB switch was set to X - Y. Replacement pair of BC178s are in position again. It seems FET and CA3046 chip were good and still are.
Additional suggestions would be much appreciated.
Best Wishes
Nigel

[LADmachining]

Hi to all.

I have had this thread bookmarked and have followed it through a number of times as I have a 465b displaying very similar symptoms to Nigel's.

The background is that I was using the 465b, that I have had for about 10 years, when I left the room briefly. Thought I heard a 'pop', then came back to find no trace on the screen where there was a signal being displayed moments previously. I could not smell any burnt components, either then or during the subsequent strip-down.

Following the advice on this thread, I checked the power supplies first. All are spot-on to their specified voltages, with less than 20mV of ripple on each.

Beam find prodices a dot on the screen that can be moved a limited amount with the vertical and horizontal position controls.

X-Y mode produces a dot that can be moved through the full extent of the screen horizontally, and about 3/4 of the screen vertically - I think this was confirmed as correct earlier in the thread.

I have found out that the scope will display a trace on the .02uS timebase setting, and trigger off a signal from the internal calibrator. It seems a bit 'touchy' though, and the trace vanishes to a shortened horizontal line, something similar to that shown in the last picture posted by Nigel, if the 'Var' control on the timebase is moved beyond a certain point. All waveforms and voltages given in the manual appear spot-on at this timebase setting.

I have removed, cleaned and tested the A-B timing board, and all values for Rt and Ct check out as far as I can tell.

I am yet to start replacing any components, but I have given all ICs and transistors a 'wiggle'.

Any advice on where to go next? I am out of my comfort zone with analogue stuff, being more used to digital (willing to learn though!).

Many thanks,

Anthony

[LADmachining]

I have managed to trace the fault I described above to a defective JFET Q4498 in the sweep generator for timebase A. It appears that this was leaky and preventing correct operation of the sawtooth generator below a certain frequency. Even when the scope was sweeping on the highest timebase setting, the actual time/div was double the value expected.

Even though the DC voltages around this part appeared correct, the fault was isolated by measuring the voltage drop across each of the resistors connected to each terminal of the FET. The voltage drop across the gate and source resistors was the same (0.202V across 100ohms) suggesting there was a gate to source leak.

The FET has been replaced and all timebase settings now function correctly. I hope this information is useful to someone else who encounters a similar problem.

Anthony

[6AL5W-Martin]

151-1025-00 I have one.

If that is the problem get it from me

greetings
Martin