Re: Hello from newcomer Fabio Trevisan - My first Tek Scope 464 + DM44
Hi David,toggle quoted messageShow quoted text
Here's some update on the H.V. transformer issue of my Tek 464.
Yesterday I finally installed the new H.V. transformer I ordered from
It was delivered 3 weeks ago to my brother in law in Toronto and last week
he came down to Brazil and brought it to me.
It came very well packed and in perfect conditions...
Since I had dismantled almost the whole H.V. circuitry, to test all the
critical components "off-board" as I did and which led me to discover that
the H.V transformer was defective, it took me a while to put everything
back together, specially because the components in this area are laid-down
in a messy mixture of P.C.B., air-mount, piggy-back and ceramic-strip
It amazes me how Tek managed to mass produce this line of 'scopes with this
kind of assembly technique.
Transformer installed and HV circuitry assembled... Turned-on the scope and
pop!... H.V. circuit's FUSE BLOWN!
First time I did it with an under rated 1A fuse for safety, instead of the
recommended 1.5A, and instead of the 4A fuse that was installed there
Then, I put a 2A fuse... and it popped! just as instantly as the 1A.
Back to square one...
Checked and rechecked all the H.V. components assembly and looked for
possible soldering shorts that could have sneaked under some component...
While keeping the fuse removed, checked all the regulated voltages... All
Then disconnected the H.V.'s drive transistor (Q1486) and measured it...
This circuit was still working when I took it apart to test the
components... And the 1A fuse blown so instantly... I don't believe that
such a short overload could have blown that massive 2N3055. The Unreg +15V
supply isn't even able to source the 15A that this beast is supposed to
withstand... And it is (was) an original TEK part!!!
I don't understand!!!
Understanding or not... Q1486 was blown... and it would be hard to source a
reliable (not fake) 2N3055 at short notice...
I managed to get an MJ15015 from a reliable supplier that I know... but I
was about the fact the the original Tek is listed as being a "SELECTED"
Installed the MJ15015, with the fuse replaced by a 20 Ohm resistor and
turned it on while measuring the drop on the resistor...
After seeing a healthy 4V drop (the 200mA draw that I once measured of this
cricuit)... I relaxed and could look at the screen...
A BEAUTIFUL TRACE WAS THERE!!! showing - as expected - a picked up 50KHz
interference, due to the H.V. cages are still not installed.
Replaced the 20 Ohm resistor by the 1A fuse and turned it back on... and NO
FUSE POPPING this time...
Still... things were not perfect. Retrace was visible... but I could make
it disappear readjusting the CRT bias (all in all, I could have touched it
while I disassembled the components all around)... But worse than the
visible retrace... the trace size (horizontal and vertical) was expanding
as I increased the "Intensity" control...
Shoot!.. Poor H.V. Regulation (or none...) I was already afraid that
something else in the circuit could have blown!!!
I went on to measure Cathode supply and...Ops!!! -1780V!!! (310V more
negative than it should)... Got suspicious that feedback loop could be
opened or that the feedback transistors could be bad...
Transistors turned out to be all good! and feedback summing point, that
should read nominally +0.4V was reading -5.2V (an indication that the
feedback divider was doing its job but even though, circuit was still not
being able to bring the output "down").
After analyzing the circuit to evaluate if it wouldn't blow anything,
decided to remove Q1484 (pulling it out of its socket) to see if the
inverter would still work (open-loop) and how much voltage it would
generate by itself (with only the baseline bias provided by R1483 of 5.6K).
It turned out that even with Q1484 removed, Cathode voltage was still at
-1780V!... meaning the the inverter was generating all that voltage without
any additional current from the feedback circuitry...Explaining why it
wasn't regulating whatsoever.
Probably, the MJ15015 has a much higher beta than the original drive
transistor...(which makes me think now that maybe Tek selected it for LOW
gain, and not HIGH).
Decided to take an intuitive shortcut and changed value of R1483, from 5.6k
to 10k... and measured cathode voltage (still in open-loop, with Q1484
Ah ha!!... much better now! Cathode voltage "dropped" to -1100V... (about
300V lower than required)...
Inserted Q1484 back and.... Voilà!!!... Cathode voltage went to -1478V...
For everything that matters, it seems to be working fine! Cathode voltage
regulates well for all CRT loads (low to high intensity and using Store
modes or not).
I don't have a Variac to allow me to check regulation for mains voltage
variation, but I did try to further increase R1483 to 18k, and H.V. still
regulates so, I think that the intuitively chosen 10k, is placing the
circuit right in the middle of its regulation range.
What do you think? Should I leave it like that, or try something else (such
as getting another transistor or dumping some of its gain stealing some of
its base current to a drain resistor...maybe to a negative rail)?
2016-10-06 1:44 GMT-03:00 David @DWH [TekScopes] <