| |
G8MNY > TECH 03.12.18 11:09l 63 Lines 3025 Bytes #999 (0) @ WW
BID : 47690_GB7CIP
Read: GUEST
Subj: Bad SMPSU design
Path: IW8PGT<IR2UBX<IW2OHX<IW0QNL<IK6ZDE<I0OJJ<GB7CIP
Sent: 181203/1004Z @:GB7CIP.#32.GBR.EURO #:47690 [Caterham Surrey GBR]
>From g8mny%gb7cip.#32.gbr.euro@i0ojj.ampr.org Mon Dec 3 11:05:14 2018
Received: from i0ojj.ampr.org by i0ojj.ampr.org (JNOS2.0k.3b) with SMTP
id AA64243 ; Mon, 03 Dec 2018 11:05:14 +0100
Message-Id: <47690_GB7CIP@gb7cip.bbs>
>From: g8mny@gb7cip.#32.gbr.euro
X-JNOS-User-Port: Telnet (gb7cip @ 82.70.39.222) -> Sending message
From: G8MNY@GB7CIP.#32.GBR.EURO
To : TECH@WW
By G8MNY (new Sep 08)
(8 Bit ASCII graphics use code page 437 or 850, Terminal Font)
Some time ago I looked at a Freeview STB that had blown up (a Digilogic
DSTB1000 which is the same as Goodmans GDB6). On first sight the SMPU design
looks fine like any other, with the normal safety features etc..
ÚÄÄÄÂÄÄÂÄÄÄÄÄÂÄÄÄÄÂÄÄÄÂÄÄÄÄÄÄÄÄÂÄÄ¿ ÚÄ´>ÃÄÂÄ +30V
_³_ _³_ ³ ³ 200k ===10n ³ ³ ||(
F /_\ /_\ ³ ³ ÀÄÄÄ´ Spike ³ )|| ÃÄ´>ÃÄÂÄ
Äo-oÂÄÄ¿ ÚÄÄÂÄÄ´ ³ ³ ³ 100 Damper³ )||(
L ³ ³ ³ ³ ³ ³ ³+ ³+ ³ ÚÄÄÄÄ)ÄÄÙ || ÃÄ´>ÃÄÂÄ +12V
³ ()) === ³ ³ === === À´<ô ³ ||(
VDR === ³ ³ ³ ³4u7 ³C _____ ³ 2M || ÃÄ´>ÃÄÂÄÄÄÄÂÄÄÂ+5V
³ ()) ³ ³ ³ ³ ³ ³ ÚÁÄÁÄÄÄÁ¿ R ||( 1k ³ ³
ÄÄÄÄÁÄÄÙ ÀÄÄÁÄÄ)ÄÄÄ´ ³ ³ ³ ) TNY265³ ³ TR ³ _³_ ³ ===
N _³_ _³_ ³ ³ ³ ÀÂÄÂÄÂÄÂÙ ³ ³opto\_/ ³ ³
/_\ /_\ ³ ³ ³ n1³ ÀÂÙ ÃÄÄÄÄ´ ³ ³ 10k ³
³D ³ ³ ³ ³ === ³ _³_Z \³< ³ \³ ³ ³
³ ³ ³ L ³ ³ ³ ³ /_\' e /³< ³ ÃÄÄ´ ³
ÀÄÄÄÁÄÄÁÄcccÄÁÄÄÁÄÄÄÁÄÄÁÄÄÁÄÄÄÄÙ ³ e/³ 1k ³
===1.5mH ÀÄÄÄÄÄÁÄÄÄÄÁÄÄÁÄ0V
WHAT WENT WRONG
The original fault was just an electrolytic "C" going open circuit, very common
& the cause of a lot of mains born RFI! But the resultant failed component
count was so high much more than the STB was worth new!!
With the SMPSU IC now having no low Z Cap C to push the current through the
transformer, all the pulse current appear as high voltage across the small
decoupling choke L (1.5mH). This bypasses the spike damper that the protects
the IC from over voltage & the IC blows up. In doing so it took out the PCB
track near the 2M R (4x 500k surface mounts) to a spare transformer primary
winding TR, the control loop protection zener Z, the L itself, 2 of the surface
mount bridge diodes D, & of course the fuses both the PCB 3A fuse & plug 5A
one!
This is relatively modern kit that has had a very short lifetime. Why do makers
make it so unreliable! The circuit component count is quite high, I can't
really see why a simple conventional mains transformer would not have been a
much better design & far more reliable!
As far as this kit goes, powering it off rather than standy will extend the
life of the caps, but repeated power on surges will stress the other
components!
See my bul on "Faulting Tips".
Why Don't U send an interesting bul?
73 De John, G8MNY @ GB7CIP
Read previous mail | Read next mail
| |
