开云体育

I've a Tek 2465B that I recently acquired, it was described as 'no trace' and that is true, when powered up various leds flash for a bit but no trace is visible, with or without the beam finder pressed. It's a 1996 model but no surface mount caps.

The first thing was to check the power supplies via the main board test point (actually a 14 pin DIL socket) J119. Measuring the voltages I get:

Pin Measured Spec
1 -14.95 -15.0
2 3.940 5.0 (digital +5v)
3 -18.66 (-15v unreg)
4 10.00 10.0
5 -4.986 -5.0
6 15.01 15.0
7 0.0 0.0 (gnd)
8 87.0 87.0
9 42.47 42.4
10 0.008 (unused)
11 -8.03 -8.0
12 0.761 5.0 (analog +5v)
13 -1.365 (?)
14 -1.363 (?)

So from this it looks like the +5v for digital is low, and the analog +5v is very low.
Looking at the PSU boards there is no evidence of either bad/leaky caps or overheated components.

Whats the next thing to test?

Hey Keith,

Do you have a second scope at hand? You probably want to next measure the
ripple on the rails, it's not unlikely that the power supply boards need
recapping.
These scopes are microprocessor controlled, and if the 5VD rail is low, the
microprocessor may never get going, or it may reset regularly. If the MPU
doesn't run, you'll have no traces, though the presence of -1.36V may
indicate that the MPU is running.

What happens when you power up the scope - you should see a sequence of
front-panel (FP) LEDs light up, then relays will click, after which the
traces should appear.
If the scope fails power-on-self-tests (POST), there will be a single FP
LED lit, and this will indicate the failing test. There is a table in the
service manual that associates FP LEDs to the failing POST. If you then
press A/B TRIG, the scope will drop into "operation" mode, and IIRC you'll
hear relays click at that point in time.

The classic problem with these scopes is electrolyte leakage on the A5
board, which will often wreck the DACs reference, which in turn will lead
to 05 POST failures and/or no traces, so that's something to look at after
you fix the PSU problems.

Good luck,
Siggi

On Wed, Dec 18, 2024 at 10:52?AM Keith via groups.io <keith=
peardrop.co.uk@groups.io> wrote:

I've a Tek 2465B that I recently acquired, it was described as 'no trace'
and that is true, when powered up various leds flash for a bit but no trace
is visible, with or without the beam finder pressed. It's a 1996 model but
no surface mount caps.

The first thing was to check the power supplies via the main board test
point (actually a 14 pin DIL socket) J119. Measuring the voltages I get:

Pin Measured Spec
1 -14.95 -15.0
2 3.940 5.0 (digital +5v)
3 -18.66 (-15v unreg)
4 10.00 10.0
5 -4.986 -5.0
6 15.01 15.0
7 0.0 0.0 (gnd)
8 87.0 87.0
9 42.47 42.4
10 0.008 (unused)
11 -8.03 -8.0
12 0.761 5.0 (analog +5v)
13 -1.365 (?)
14 -1.363 (?)

So from this it looks like the +5v for digital is low, and the analog +5v
is very low.
Looking at the PSU boards there is no evidence of either bad/leaky caps or
overheated components.

Whats the next thing to test?

Thanks Siggi, I'll put a scope on the rails when I've managed to put the PSU boards back in!
I did have a good visual check of them and there is no obvious electrolyte leakage or bulging, but while out they may benefit from a recapping.
With the +5V digital rail at less than 4v, I doubt if any logic is working although there may be massive ripple that my dmm is averaging out.

How on earth do you get the PSU boards back in though? The wires from the voltage selector switch are short and it seems can only be reconnected when the board is in situ, but then they are underneath the line filter and unreachable - is there some trick to reconnecting them?

Keith,

Needle nose pliers to put the connectors back on that are under the mains filter. The power switch mechanics at the rear going to the board must be in the right position for the board to go back fully and properly.

Also, get rid of those RIFAs! I put in X1 and Y1 types as replacments. There are the two larger ones off the mains that are the first to go. There are others elsewhere in the primary of the power supply. You can use 630V film types for some if the X1 or X2 is not made in those values. I used higher voltage types for the ones off the mains. I think it is 310V or more in mine. Higher voltage is fine.

The recapping can be done by a kit sold by a member here, Condor Audio. You can buy the caps yourself. If the latter, get high temp., low ESR and long life. Increase the capacitance. The kit also has some resistors that should be replaced. There are several resistors elsewhere that would be best to replace. Three are 1/4W carbons that will be way high and several on the high voltage board that are stressed in operation. Go to 1W for these.

Mark

On Wed, Dec 18, 2024 at 2:04?PM Keith via groups.io <keith=
peardrop.co.uk@groups.io> wrote:

How on earth do you get the PSU boards back in though?

I'd use a VERY small crowbar :).

The wires from the voltage selector switch are short and it seems can only
be reconnected when the board is in situ, but then they are underneath the
line filter and unreachable - is there some trick to reconnecting them?

It's been a long time since I did this, but I seem to remember using long
nose pliers to connect them on the PSU side.

I do think, however, that there must exist people with at least an extra
joint on at least one arm. Whether it's an elbow or a wrist, I don't know.
What I do know is that I bled getting the shorted "blower resistor" out of
my Volvo, and I don't see how I would have been able to fit the new one
given my anatomical arrangement.

PS: A modern "blower resistor" is a resistor only in naming legacy.

On Wed, Dec 18, 2024 at 08:04 PM, Keith wrote:

How on earth do you get the PSU boards back in though? The wires from the
voltage selector switch are short and it seems can only be reconnected when
the board is in situ, but then they are underneath the line filter and
unreachable - is there some trick to reconnecting them?

Loosening the appropriate screws on the rear panel/back plate will release it, making life a lot easier. Also, ISTR that just loosening the screws holding the line filter to the back plate helps.

As regards the 4VDC where it should be 5VDC, just replace the buffer caps. Often, a large drop in capacity in either or both of the caps in the pi-filters results in reasonably flat but too low voltages. You should release all electrolytic caps in the PSU anyway. Replace them one by one, paying attention to their polarity because some boards are known to have errors in their printing.

Raymond

On Wed, Dec 18, 2024 at 09:48 PM, Raymond Domp Frank wrote:

You should release all electrolytic caps

That should have read "You should *replace* all electrolytic caps...

Raymond

On Wed, Dec 18, 2024 at 3:48?PM Raymond Domp Frank via groups.io <hewpatek=
gmail.com@groups.io> wrote:

Replace them one by one, paying attention to their polarity because some
boards are known to have errors in their printing.

^
|
This!

Keith, you're not exactly new here, but in this group we see a lot of
repeats and we tend to go a little bit "case blind".
Raymond and Mark have the good sense to point out the most common PSU
gotchas from the "lore" that you may not be aware of until you have an
actual case study :).

Some good tips here, thank you folks. I'll start making up an order for DigiKey...

May I ask one more question: Can the PSU board(s) be powered up without their outputs driving the rest of the scope - are load resistors needed for the psu to work standalone? As its going to save time testing if I don't have to keep taking the board out to change a cap then put it back in to test (the suggustion to change them once at a time is very wise).

Keith

?ann mi?., 18. des. 2024 kl. 16:41 skrifa?i Keith via groups.io <keith=
peardrop.co.uk@groups.io>:

May I ask one more question: Can the PSU board(s) be powered up without
their outputs driving the rest of the scope - are load resistors needed for
the psu to work standalone?

The service manual describes the necessary loads for the PSU to run the
boards standalone. You can run the inverter board alone with a single load,
but the LV board needs a few. It’s probably sufficient for regulation to
load only the 5VD, though of course you won’t know whether the other
supplies will sustain a load.

Keith,

If you want, I can send a photo of the parts list I used for mine that has Mouser part numbers. The list is several years old so prices and parts may not be the same.

The two main filters that are axial, I used Nichicon LGR types that are 390mfd than added wires from the lugs to the board. Measure the diameter in mm to make sure the size you get is going to physically fit. If yours has the resistor across the NTC in the mains, that can be removed. Higher resistance means lower inrush current. Not all of these have the resistor in parallel with the NTC.

There are two 4,7mfd tantalums that are known to go open. The kit I mentioned has film types to replace them. Replace these tantalums on sight.

I did mount the replacement electrolytics off the board some to allow air flow under.

There could/likely be a mistake of the marking of two electrolytics on the board. Replace them using the polarity the originals came out. The marking differs from the schematic.

The two 39meg carbons can be replaced with a pair of 20meg in series. The additional meg will not hurt. The 100meg carbon will be way high (maybe 160meg).

Oil the fan.

With mine, I put a right angle shielded cord on it. I will help keep noise from getting in or out. I also used some MX-4 compound on the bottom of the heatsink that is flat to the inside metal to aid in heat transfer. This part I did last before installing the power supply boards then bolting it down.

Mark

See EEVBLOG, 2465B teardown, 72 PAGES of gerat info.

See Condor Audio , Menecham in Tel-Aviv has PSU recap kits.

See A5 control board leaking SMD lytics for the complete A5 control board destruction by corrosion an drepair, restore tips and proceedue.

If the A5 is the newer SN with the Dallas NVRAM and SMD lyticsyou WILL have a bad control /DAC /REF that MUST be fixed before anyting will work.

Also see Tekscopes2, similar but DIFFERET group.io.

Bon chance


Jon

On Wed, Dec 18, 2024 at 10:48 PM, Mark Vincent wrote:

If you want, I can send a photo of the parts list I used for mine that has
Mouser part numbers. The list is several years old so prices and parts may not
be the same.

Actually that would be very useful, for checking I haven't missed any.

cheers

Keith

Keith,

I posted my older list to the photos section. The 100meg can be the radial leaded blue type found on Mouser. I think it is a 3/4W that will directly fit in the original holes.

Mark

Keith,

For the two 39meg resistors (R710 and R910 on page <5>), two of 594-VR37000002005FA1 in series will work. For the 100 meg (R1991 on page <8>), 708-HVA12FA100M or 588-SM104031006FE (check length of pins of original and this on to make sure it will drop in without bending the leads. I looked under 'ohmite resistor' then 'thick film resistor'). The part numbers are Mouser. If you have any more questions, I will be glad to help. If you chose to raise the NTV value to 10/16 ohms, a 527-CL60/70 will work in place of RT1010 off the mains.

Depending upon the amount of dust inside, removing the crt then cleaning it and the shield may be best. Removing the dust prevents a leakage path due to dust and moisture on the dust. Clean the filters also while the crt is out. That will make the trace clearer and sharper due to dirt being removed.

Mark

Thanks Mark, that's very helpful.

Keith

Keith,

The NTV should be NTC.

If you have the memoury backup battery in a brick shape, 667-TL5903P is what I put in mine. It is axial leaded. I mounted the battery a bit off the board while bending the leads to fit in the holes. This current capacity is higher than original and costs less than one in the original shape.

Mark

So I finally received the capacitors I ordered from DigiKey on 1st Jan, apparently there was an IT issue that caused the order to be delayed, then the weather affected shipping... Anyhow, I replaced all the electrolytics on the A2A1 and A3 boards, plus the rifa's on the A3 inverter board.

Made up a load board with a 2.2ohm 20W for the +5VD, and 100ohm 5W resistors for the +5V, -5V, +15V, -15V, -8V. But 8.2K for the 84V and 4.3K for 42V rails as it seemed the current draw would be a bit OTT for those rails with 100ohms.

Anyhow testing the board with that load looked good, voltages were what they should be, so I fitted the PSU back. Measured J119 pins are now:

pin1 (-15v) = -14.957
pin 2 (+5V digital) = 5.006
pin 3 = (-15v unreg) -18.855
pin 4 (10v ref) = 10.004
pin 5 (-5v) = -4.985
pin 6 (+15v) = 15.014
pin 7 (gnd) = -0.001
pin 8 (87v) = 87.02
pin 9 (42v) = 42.46
pin10 (unused) = 0.004
pin 11 (-8v) = -8.032
pin 12 (5v analog) = 4.971
pin 13 = -1.3705
pin 14 = -1.3692

On power up, I got a test fail: 'Test 04 Fail 12 TV<6400>'. From what I read this is the Dallas Semi NVRAM cal. The datecode on that chip is '96 so I guess it needs changing?

The readout of e.g. ch1, ch2, timebase seems correct but has '?' in all the spaces... does this happen because the cal is invalid?

Lastly, there seems to be a triggering problem. On slow X axis speeds e.g. 200mS/cm I can get a correctly triggerred trace, but on faster ones the triggering either is intermittent or fails completely. It's not clear if this is an intermittent connection or something else. Any thoughts?

cheers,

Keith

Keith

The error message indicate dead NVRAM battery

Needs ver careful removing, place low profile machine pin 28 pin socket, get new genuine Dallas Semi replacement, use EEPROM programming to load default CAL


Perform a complete CAL according to the service manual.

All other faults....check PS and HV for failed caps, check A5 control board for leaks of SMD lyrics.

See both Tektronix forums on groups.io and EEVblog 2465 teardown thread.

Also Condor Audio and hams for 2465 rebuild notes

Bon chance


Jon

Keith,

I just notice this statement in your first post: "It's a 1996 model but no surface mount caps."
Are you saying the A5 controller is not using surface mount parts?

--Victor