Re: 465B won't trigger

Fabio Trevisan
 

Hi Lorn,

You're right... 2N5245 is indeed recommended as a replacement to 2N4416 and, as Chuck mentioned, the 2N4416's 4th pin can be cut-out. Just don't cut the wrong one.
I`m looking right now at a data-sheet page from NewJersey Semiconductor products:
http://pdf1.alldatasheet.com/datasheet-pdf/view/459118/NJSEMI/2N5245.html
Only difference that I see that really screams to the eye is the Gate leakage current, it's 1nA on the 2n2545, against 0.1nA on the 2n4416... but either one would not cause problems on the trigger input buffer (worst case 1nA on 1Mohm produces only 1mV error).
I think you can be happy with either one, whichever you can find.
When I found mine with problem, and before I got a pair of original Tek parts from a fellow from this group (from whom I bought a few other parts I was needing back then), I also found a handful of 2n4416 and selected the pair that would show the smallest DC error.
The good thing about this input circuit is that you can use the circuit itself to match the transistors... Just swap them through all combinations and select the pair that produces the smallest DC voltage (having the input selected to External, DC, and having the BNC input shorted to ground).

Rgrds,

Fabio

On Fri, Apr 27, 2018 at 07:23 am, blackholenulled wrote:


Fabio:
Yeah, I was hoping for "been there, done that", so you've been a big help!

So this is interesting what you're saying about the cross reference. The
transistors I pulled out are 3 pin TO-92 case devices. The data sheets I'm
finding for 2N4416 show it as a four pin device. I wonder if there is a three
pin variant?

I need to compare the data sheets but there are indications from other sources
that 2N4416 and 2N5245 are electrically equivalent.

So how I arrived at 2N5245:

https://www.wbparts.com/rfq/5961-00-436-3317.html

Which lists as a cross reference 2N5245. It came from military specs, so I
figure if it's good enough for the military, it's good enough for my purposes.

This discussion indicates 2N5245 and 2N4416 are equivalent:
https://www.vintage-radio.net/forum/showthread.php?t=67673

I figure if I can't purchase a matched pair I can buy a handful and do some
matching using this article:
http://www.firstwatt.com/pdf/art_matching.pdf

So the Tektronix cross reference you linked is a great resource. I'm just
puzzled as to why Tektronix would list a four pin device rather than three
pin, unless it turns out there is a three pin variant of 2N4416 that used to
be available. (well, quick search on the internet indicates there are TO-92
case variants, but more common is TO-72 four pin for the 2N4416).

What I need to do is remove the heat sink from the transistor pair and see
what the markings are.

Lorn
KK4KRI

On Thu, Apr 26, 2018 at 01:31 pm, Fabio Trevisan wrote:


Hello Lorn,
Great news indeed!... So your problem is narrowed down to exactly the same
cause of mine: One, or both, of the input FET Transistors bad.
And I did exactly what you did, meanwhile I couldn't source another pair...
I
cannibalized the B sweep trigger.
In fact, there's very little impact to it, because 90% of the times, we use
the B Sweep in the "STARTS AFTER DELAY" mode... and this mode doesn't use
the
triggering input buffers or the trigger amplifier whatsoever...

As for the source of the transistors you wrote:
Any ideas anyone?
Manufacturer part number: SF50031, which comes back to 2N5245 N channel
JFET, which is obsolete. It's late and I haven't tried to cross reference
that
yet.

Not quite right!
I don't know from where you got the 2N5245, but according to
Tektronix_Xref_sm.pdf available at Archive.org
(https://archive.org/details/tektronix_Tektronix_Xref_sm), part number
151-0042-00 is found at page 7-7.
This part number is a characterized and matched pair of 2N4416 (which is
probably also obsolete, but not that hard to find).
From the "stock" 2N4416 whose IDss can vary from 5 to 15 mA, those ones are
selected to a narrower IDss range of 10 to 15mA, and the pair need to be
matched within 0.5mA between each other.

Even if you cannot get a matched pair, as long as they're not too different,
you can zero them out by tweaking the value of R7325 or R7327 (or both).
This will compensate for the different Vgs that will develop across each one
of the **not matched** xsistors' source resistors, in order to get the
output
voltage as close to 0V as possible,
The fact they're not matched will have the down side their IDss will also
drift differently with temperature so, tweaking the resistors solves the
matching problem at one given temperature but as the temperature changes,
they
will drift apart from each other...
But since this is **only** triggering (I can see Howard Vollum revolving in
his grave now), it's not that critical!. And you can always put the
non-matched pair on the B sweep trigger, which is rarely used.

Good luck with your 465B, This is my dream 465 scope, with Alternate Delayed
time base instead of the least useful "MIX" mode that's present on the
regular
465s.
I have a 464 with DM44 multimeter (it has a weak CRT) and just recently got
a
7623A with 7B53A timebase, so I'm probably married to the "MIX" mode for the
foreseeable future.
I'll probably sell my 464 and watch out for a good 465b, or even a 468
(which
is a 465b + digital storage).

Rgrds,

Fabio

Join TekScopes@groups.io to automatically receive all group messages.