Re: tek 2246 Mod A SMPS failure


Priya Jakatdar
 

I found the problem - the switching transistors Q2209/10 alternatively pulling current from the 44 VDC rail to the switching power FET are shorted out at their C-E junction. To further validate my theory, I injected a 2V constant current power supply signal into the 44 VDC rail and it is completely shorted to the Q2214 FET Drain, pulling as much current as I will allow - I stopped at 100 mA not wanting to damage anything. The CE voltage was just 20 mV at that point.

Looks like Tek realized that this is a recurring issue because my SMPS has a jumper between the Q2209/2210 emitters and the Q2214 Drain as if to almost suggest to try cutting that first when there are SMPS issues. I will do that to verify the issue before thinking about repairs. I am also looking for some way to verify that the Q2214 FET and T2204 transformer work before I proceed further.

I wonder if a fusible link or 3A fast burn fuse would work in place of the jumper, so these transistors will be protected.

More to come,

Priya.

--- In TekScopes@yahoogroups.com, "Mover" <priya_jakatdar@...> wrote:

I separated out the SMPS from the main board and measured resistance.

(1) None of the SMPS diodes are shorted or open.
(2) None of the SMPS transformer secondary windings are shorted or open.
(3) None of the SMPS low voltage supply capacitors are shorted. I am looking around to see if any of my buddies has a in-circuit capacitance ESR tester, but that appears to be a long shot given that the SMPS failure occurred following a main board chip failure.
(4) The resistance measurements are fine and shown below in sequence (Both, Board alone, SMPS alone) for each supply line. HIGH means over 5 Mohm. No obvious shorts anywhere.
(a) +5V (95 ohm, 95 ohm, HIGH)
(b) line-trig (77 ohm, HIGH, 77 ohm)
(c) -5V (41 ohm, 41 ohm, HIGH)
(d) -15V (575 ohm, 750 ohm, 2300 ohm) - 2300 ohm fan in the SMPS.
(e) +15V (480 ohm, 583 ohm, HIGH)
(f) +7.5V (134 ohm, 134 ohm, HIGH)
(g) -7.5V (143 ohm, 143 ohm, HIGH)
(h) +55V (HIGH, HIGH, HIGH)
(i) +130V(10 Kohm, 10 Kohm, HIGH)
(j) 130VAC (2.9 ohm, HIGH, 2.9 ohm) - transformer winding
(5) I applied 5V to the main board with a Kikusui Constant Current DC power supply, gradually increasing the current limit to 100 mA on each of the supplies. The 5V and -5V lines draw increasing current and the supply voltage rises nicely and smoothly from 0 to 3V as I allow up to 100 mA of current. The other load lines hardly draw any current at all at 5V. No hard short on the main board would exhibit this pattern.

I am back to thinking this is an internal SMPS issue where I am too fearful of the high voltages to attempt anything more than a resistance check. I will post my resistance measurements summary with any findings.

As always, I appreciate any and all input from the community on this.

Priya.

--- In TekScopes@yahoogroups.com, "Rob" <rgwood@> wrote:

"Thanks for helping me think through an alternative approach."

Tis indeed the reason I remain a highly satisfied and sometimes active forum
member. It is like a little family where everyone has a unique perspective
and I enjoy them all..

Your welcome and Thank you back for understanding my "come from".
Rob

-----Original Message-----
From: TekScopes@yahoogroups.com [mailto:TekScopes@yahoogroups.com] On Behalf
Of Mover
Sent: Sunday, December 02, 2012 1:39 PM
To: TekScopes@yahoogroups.com
Subject: [TekScopes] Re: tek 2246 Mod A SMPS failure

Hi Rob,

I had measured resistance earlier at all secondary supplies and within the
SMPS and they appeared to be OK, offering no clue. However, in circuit
resistance measurements can be misleading as noted in other posts.

You now make two great points neither of which require the firing up of the
SMPS: (1) Measure main board and SMPS resistance after disconnecting the
SMPS and (2) Inject external power into the main board to troubleshoot it
further should it have a short.

I will do both and report findings without messing with the SMPS and the
risk of the high voltages especially with the CRT filament and anode.

Thanks for helping me think through an alternative approach.

Priya.

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