tektronix 2235 , blown switchmode

[john]

Hi All!

I have an old 2235 'scope which has died on me.
It looks like the resistor R909 and Q908 are fried . It also looks like
the IRF820 ( on heatsink) is blown.
Can anyone help with an excerpt from a service manual ( diag) or part
numbers/values ?

Does this happen often with these babies ?

Regards

John

--
Using Opera's revolutionary e-mail client: http://www.opera.com/m2/

 

[Jules]

"john" <john@agent-j.com> wrote

I have an old 2235 'scope which has died on me.
It looks like the resistor R909 and Q908 are fried . It also looks like
the IRF820 ( on heatsink) is blown.
Can anyone help with an excerpt from a service manual ( diag) or part
numbers/values ?

Does this happen often with these babies ?

Oh Heck! - I hope not, always fearful for mine; but I did manage to get the
manual - will mail you the "Power Supply & CRT" schematic. Does anyone else
have experience of this fault?
- Julian

 

[Jim Yanik]

john <john@agent-j.com> wrote in
newspsa7aefh90tb1i5@funkmeister.agent-j.com:

Hi All!

I have an old 2235 'scope which has died on me.
It looks like the resistor R909 and Q908 are fried . It also looks
like the IRF820 ( on heatsink) is blown.
Can anyone help with an excerpt from a service manual ( diag) or part
numbers/values ?

Does this happen often with these babies ?

Regards

John

here's some data I compiled while at TEK;

TO-220 FET SPECS


+
151-1136-00 MTP12N10E 100V 0.16ohm 14.0A
151-1141-00 IRF730,STP300H 400V 1.0 ohm 5.5A
151-1151-00 IRF710/MTP3N40 400V 3.6 ohm 1.5A
151-1152-00 IRF820/MTP475 500V 3.0 ohm 2.5A
151-1171-00 BUZ71A/MTP15N05E 50V 0.12ohm 14 A
151-1214-01 IRF830 500V 1.5 ohm 4.5A
151-1245-00 MTP6N60E 600V 1.2 ohm 6.0A
151-1286-00 MTP4N80E/BUK456-800 800V 3.0ohm 4.0A



The Molex connector to the pre-reg FET(IRF-820) should be removed and the
leads soldered direct to the FET,that's a TEK 'mandatory' mod(meaning it
shoulda been done when in for any service).The Molex would discolor and
burn up.

R909 is a 1 ohm,ww,fusible 2W resistor,p/n 308-0677-00
I believe Q909 is a TIP31A,and you should replace both of the pair when one
blows.(TEK p/n 151-0476-02

You should also check the PS electrolytics for bad ESR.
The rectifier diodes also may short.

Watch out for poor focus/astig,the focus string resistors increse in value
and eventually open and should be replaced with carbon composition types
instead of the carbon film types in there now.

The way this PS works is that the line V is rectified and filtered,about
170VDC,and then a switching pre-regulator(TL494/594 based) drops it to
~43VDC,then the power oscillator converts that to the secondary voltages
and HV.
--
Jim Yanik
jyanik-at-kua.net

 

[john]

Hi Jim,
thanks to both you and Jules for the fast response
From what i see, the tranny (Q908)thats blown has completely blown its top
off and i can understand why...
The fet blew, shorting gate to drain... hence all the current went through
this poor little guy ( Q908, to-92 package).. frying it, the diode CR908
and R909.

the Snippet sent to me by Julian, has no R909, which is whats written on
my PCB and therefore i think there are one or two version differences, as
mine is the 220V version.
The remaining band on the resistor is an orange ( 3xxxx) which makes me
think it is a 3k.

any more clues as i'm going to get the parts now and replace and then fire
up



On Fri, 16 Jul 2004 00:16:25 +0000 (UTC), Jim Yanik <jyanik@abuse.gov>
wrote:

john <john@agent-j.com> wrote in
newspsa7aefh90tb1i5@funkmeister.agent-j.com:


Hi All!

I have an old 2235 'scope which has died on me.
It looks like the resistor R909 and Q908 are fried . It also looks
like the IRF820 ( on heatsink) is blown.
Can anyone help with an excerpt from a service manual ( diag) or part
numbers/values ?

Does this happen often with these babies ?

Regards

John


here's some data I compiled while at TEK;

TO-220 FET SPECS

+
151-1136-00 MTP12N10E 100V 0.16ohm 14.0A
151-1141-00 IRF730,STP300H 400V 1.0 ohm 5.5A
151-1151-00 IRF710/MTP3N40 400V 3.6 ohm 1.5A
151-1152-00 IRF820/MTP475 500V 3.0 ohm 2.5A
151-1171-00 BUZ71A/MTP15N05E 50V 0.12ohm 14 A
151-1214-01 IRF830 500V 1.5 ohm 4.5A
151-1245-00 MTP6N60E 600V 1.2 ohm 6.0A
151-1286-00 MTP4N80E/BUK456-800 800V 3.0ohm 4.0A



The Molex connector to the pre-reg FET(IRF-820) should be removed and the
leads soldered direct to the FET,that's a TEK 'mandatory' mod(meaning it
shoulda been done when in for any service).The Molex would discolor and
burn up.

R909 is a 1 ohm,ww,fusible 2W resistor,p/n 308-0677-00
I believe Q909 is a TIP31A,and you should replace both of the pair when
one
blows.(TEK p/n 151-0476-02

You should also check the PS electrolytics for bad ESR.
The rectifier diodes also may short.

Watch out for poor focus/astig,the focus string resistors increse in
value
and eventually open and should be replaced with carbon composition types
instead of the carbon film types in there now.

The way this PS works is that the line V is rectified and filtered,about
170VDC,and then a switching pre-regulator(TL494/594 based) drops it to
~43VDC,then the power oscillator converts that to the secondary voltages
and HV.

--
Using Opera's revolutionary e-mail client: http://www.opera.com/m2/

 

[Jim Yanik]

john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

Hi Jim,
thanks to both you and Jules for the fast response
From what i see, the tranny (Q908)thats blown has completely blown
its top
off and i can understand why...
The fet blew, shorting gate to drain... hence all the current went
through this poor little guy ( Q908, to-92 package).. frying it, the
diode CR908 and R909.

the Snippet sent to me by Julian, has no R909, which is whats written
on my PCB and therefore i think there are one or two version
differences, as mine is the 220V version.
The remaining band on the resistor is an orange ( 3xxxx) which makes
me think it is a 3k.

Well,I'm working from my 2213/15 manual,but the PS circuitry is almost
identical. There mare 47 ohm Resistors in the bases of the power osc
xstrs,those can open when one of the xstrs fails.The 1 ohm ww resistor I
mentioned is in the emitter of both the power osc xstrs.It is very common
to fail.
Perhaps your crowbar circuit is what failed.
This is an overvoltage circuit that is supposed to put the pre-reg into
'burst' mode(current limit) by drawing a large current thru the SCR.

There's a 3 ohm,3w,5% ww R in that circuit along with a SCR (Q935,151-0506-
00,C106B2X283 mfg p/n) and a 51V zener,0.4w,5%.

any more clues as i'm going to get the parts now and replace and then
fire up



On Fri, 16 Jul 2004 00:16:25 +0000 (UTC), Jim Yanik <jyanik@abuse.gov
wrote:

john <john@agent-j.com> wrote in
newspsa7aefh90tb1i5@funkmeister.agent-j.com:


Hi All!

I have an old 2235 'scope which has died on me.
It looks like the resistor R909 and Q908 are fried . It also looks
like the IRF820 ( on heatsink) is blown.
Can anyone help with an excerpt from a service manual ( diag) or
part numbers/values ?

Does this happen often with these babies ?

Regards

John


here's some data I compiled while at TEK;

TO-220 FET SPECS

+
151-1136-00 MTP12N10E 100V 0.16ohm 14.0A
151-1141-00 IRF730,STP300H 400V 1.0 ohm 5.5A
151-1151-00 IRF710/MTP3N40 400V 3.6 ohm 1.5A
151-1152-00 IRF820/MTP475 500V 3.0 ohm 2.5A
151-1171-00 BUZ71A/MTP15N05E 50V 0.12ohm 14 A
151-1214-01 IRF830 500V 1.5 ohm 4.5A
151-1245-00 MTP6N60E 600V 1.2 ohm 6.0A
151-1286-00 MTP4N80E/BUK456-800 800V 3.0ohm 4.0A



The Molex connector to the pre-reg FET(IRF-820) should be removed and
the leads soldered direct to the FET,that's a TEK 'mandatory'
mod(meaning it shoulda been done when in for any service).The Molex
would discolor and burn up.

R909 is a 1 ohm,ww,fusible 2W resistor,p/n 308-0677-00
I believe Q909 is a TIP31A,and you should replace both of the pair
when one
blows.(TEK p/n 151-0476-02

You should also check the PS electrolytics for bad ESR.
The rectifier diodes also may short.

Watch out for poor focus/astig,the focus string resistors increse in
value
and eventually open and should be replaced with carbon composition
types instead of the carbon film types in there now.

The way this PS works is that the line V is rectified and
filtered,about 170VDC,and then a switching pre-regulator(TL494/594
based) drops it to ~43VDC,then the power oscillator converts that to
the secondary voltages and HV.

Your scope should regulate from 90-250 VAC without any internal changes.

--
Jim Yanik
jyanik-at-kua.net

 

[TekMan]

Jim Yanik <jyanik@abuse.gov> wrote in message news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...

john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

Hi Jim,
thanks to both you and Jules for the fast response
From what i see, the tranny (Q908)thats blown has completely blown
its top
off and i can understand why...
The fet blew, shorting gate to drain... hence all the current went
through this poor little guy ( Q908, to-92 package).. frying it, the
diode CR908 and R909.


snipp..
151-1136-00 MTP12N10E 100V 0.16ohm 14.0A
151-1141-00 IRF730,STP300H 400V 1.0 ohm 5.5A
151-1151-00 IRF710/MTP3N40 400V 3.6 ohm 1.5A
151-1152-00 IRF820/MTP475 500V 3.0 ohm 2.5A
151-1171-00 BUZ71A/MTP15N05E 50V 0.12ohm 14 A
151-1214-01 IRF830 500V 1.5 ohm 4.5A
151-1245-00 MTP6N60E 600V 1.2 ohm 6.0A
151-1286-00 MTP4N80E/BUK456-800 800V 3.0ohm 4.0A



snip..

Well, do as Jim proposes.
Followind steps are helpful if you have replaced all other parts the
other posters have mentioned.

You can check the circuitry following the hv smps simply by powering
the scope from a regulated laboratory DC supply: Locate the 1000 mikro
F (uF) cap on the board, apply approx. 45 v Dc from the lab supply /
1.5 A . current limit is very useful (CC/CV-mode lab grade supply).

Scope should power up now and be usable partly. HINT: No mains
connection for this trobuleshoouting, so unpluq the mains lead from
the scope!! Otherwise, you'll fry yourself or other pars ;-).

Be careful not to get more than 49 v DC in this step (or the crowbar
circuit will trigger).

If you have excessive current, check also the flyback diode in the
primary SMPS circuit. It's likely to blow, and causes a nasty short
circuit for the 1000 uF cap. A 200V 2 A fast switch ( < 80 nanosec)
diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be replaced
with a TL 494 ( the 594 is a little bit harder to get, at least here
on the european continent).

hth,
andreas

 

[Jim Yanik]

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407190004.a0b4d86@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...
john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

Hi Jim,
thanks to both you and Jules for the fast response
From what i see, the tranny (Q908)thats blown has completely blown
its top
off and i can understand why...
The fet blew, shorting gate to drain... hence all the current went
through this poor little guy ( Q908, to-92 package).. frying it,
the diode CR908 and R909.


snipp..

Well, do as Jim proposes.
Followind steps are helpful if you have replaced all other parts the
other posters have mentioned.

You can check the circuitry following the hv smps simply by powering
the scope from a regulated laboratory DC supply: Locate the 1000 mikro
F (uF) cap on the board, apply approx. 45 v Dc from the lab supply /
1.5 A . current limit is very useful (CC/CV-mode lab grade supply).

Scope should power up now and be usable partly. HINT: No mains
connection for this trobuleshoouting, so unpluq the mains lead from
the scope!! Otherwise, you'll fry yourself or other pars ;-).

Be careful not to get more than 49 v DC in this step (or the crowbar
circuit will trigger).

If you have excessive current, check also the flyback diode in the
primary SMPS circuit. It's likely to blow, and causes a nasty short
circuit for the 1000 uF cap. A 200V 2 A fast switch ( < 80 nanosec)
diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be replaced
with a TL 494 ( the 594 is a little bit harder to get, at least here
on the european continent).

hth,
andreas

The crowbar circuit is what was fried on his unit,I believe.
When the FET shorted,the full line V was applied to the crowbar.

I surmise that Q908 is the SCR,CR908 is the zener,and R909 is the 3 ohm
fusible ww reistor.
And yes,the TL594/494 could have blown along with the FET.

--
Jim Yanik
jyanik-at-kua.net

 

[TekMan]

Jim Yanik <jyanik@abuse.gov> wrote in message news:<Xns952B71ED9A153jyanikkuanet@204.117.192.21>...

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407190004.a0b4d86@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...
john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

snipp..
Be careful not to get more than 49 v DC in this step (or the crowbar
circuit will trigger).

If you have excessive current, check also the flyback diode in the
primary SMPS circuit. It's likely to blow, and causes a nasty short
circuit for the 1000 uF cap. A 200V 2 A fast switch ( < 80 nanosec)
diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be replaced
with a TL 494 ( the 594 is a little bit harder to get, at least here
on the european continent).

hth,
andreas


The crowbar circuit is what was fried on his unit,I believe.
When the FET shorted,the full line V was applied to the crowbar.

I surmise that Q908 is the SCR,CR908 is the zener,and R909 is the 3 ohm
fusible ww reistor.
And yes,the TL594/494 could have blown along with the FET.

Hi Jim,

well, you're right: In most of the 2235 I repaired with a fried smps
the fast switch diode is OK. I only had it once in a case of 15
broken 2235's I had to repair that the fast switch diode was blown ( a
dead short diode). This was a sour job: It took a while to diagnose
this, but finally turned out.

Sympthoms were: blown fuse, dead IC, dead FET and dead fast switch.
Crowbar was intact and OK.

Wel, but after rethinking this: The problem can be more clever
diagnosed when a variac is handy: With a good fuse (and diagnosed GOOD
crowbar) and the dead short FET in the scope, scope should power up
partly when one raises AC inlet voltage from 0 to 30V AC. If it powers
up, the fast switch diode is ok with confidence 99.9%. If not: I'd
recommend you check the diode closely.

Yes, one could use a curve tracer too for powering the scope - if
you're lucky to have one ;-)

hth,
Andreas

 

[TekMan]

Jim Yanik <jyanik@abuse.gov> wrote in message news:<Xns952B71ED9A153jyanikkuanet@204.117.192.21>...

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407190004.a0b4d86@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...
john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

snipp..
Be careful not to get more than 49 v DC in this step (or the crowbar
circuit will trigger).

If you have excessive current, check also the flyback diode in the
primary SMPS circuit. It's likely to blow, and causes a nasty short
circuit for the 1000 uF cap. A 200V 2 A fast switch ( < 80 nanosec)
diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be replaced
with a TL 494 ( the 594 is a little bit harder to get, at least here
on the european continent).

hth,
andreas


The crowbar circuit is what was fried on his unit,I believe.
When the FET shorted,the full line V was applied to the crowbar.

I surmise that Q908 is the SCR,CR908 is the zener,and R909 is the 3 ohm
fusible ww reistor.
And yes,the TL594/494 could have blown along with the FET.

Hi Jim,

well, you're right: In most of the 2235 I repaired with a fried smps
the fast switch diode is OK. I only had it once in a case of 15
broken 2235's I had to repair that the fast switch diode was blown ( a
dead short diode). This was a sour job: It took a while to diagnose
this, but finally turned out.

Sympthoms were: blown fuse, dead IC, dead FET and dead fast switch.
Crowbar was intact and OK.

Wel, but after rethinking this: The problem can be more clever
diagnosed when a variac is handy: With a good fuse (and diagnosed GOOD
crowbar) and the dead short FET in the scope, scope should power up
partly when one raises AC inlet voltage from 0 to 30V AC. If it powers
up, the fast switch diode is ok with confidence 99.9%. If not: I'd
recommend you check the diode closely.

Yes, one could use a curve tracer too for powering the scope - if
you're lucky to have one ;-)

hth,
Andreas

 

[Jim Yanik]

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407202154.764c02bf@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns952B71ED9A153jyanikkuanet@204.117.192.21>...
and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407190004.a0b4d86@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...
john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

snipp..
Be careful not to get more than 49 v DC in this step (or the
crowbar circuit will trigger).

If you have excessive current, check also the flyback diode in the
primary SMPS circuit. It's likely to blow, and causes a nasty short
circuit for the 1000 uF cap. A 200V 2 A fast switch ( < 80
nanosec) diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be
replaced with a TL 494 ( the 594 is a little bit harder to get, at
least here on the european continent).

hth,
andreas


The crowbar circuit is what was fried on his unit,I believe.
When the FET shorted,the full line V was applied to the crowbar.

I surmise that Q908 is the SCR,CR908 is the zener,and R909 is the 3
ohm fusible ww reistor.
And yes,the TL594/494 could have blown along with the FET.

Hi Jim,

well, you're right: In most of the 2235 I repaired with a fried smps
the fast switch diode is OK. I only had it once in a case of 15
broken 2235's I had to repair that the fast switch diode was blown ( a
dead short diode). This was a sour job: It took a while to diagnose
this, but finally turned out.

Sympthoms were: blown fuse, dead IC, dead FET and dead fast switch.
Crowbar was intact and OK.

Wel, but after rethinking this: The problem can be more clever
diagnosed when a variac is handy: With a good fuse (and diagnosed GOOD
crowbar) and the dead short FET in the scope, scope should power up
partly when one raises AC inlet voltage from 0 to 30V AC. If it powers
up, the fast switch diode is ok with confidence 99.9%. If not: I'd
recommend you check the diode closely.

Yes, one could use a curve tracer too for powering the scope - if
you're lucky to have one ;-)

hth,
Andreas

I don't think the pre-reg will fire up at such a low line V. There's a
couple of xstrs that supply power to the TL594 pin 12 that act as a delay
to allow the input line filter cap to charge up before the PWM IC starts.It
will also act as an undervoltage lockout.(normal op range is 90-250 VAC)
IIRC,line VAC has to get up to 60 VAC or greater for the IC to start.
One can hook up a LV ps across the "housekeeping cap" for the TL594 and
slowly raise the V to ~16 VDC and the IC will start,and a pulse train can
be scopes at pin10 or a ramp at pin5.

With the scope NOT connected to mains!!!


--
Jim Yanik
jyanik-at-kua.net

 

[TekMan]

Jim Yanik <jyanik@abuse.gov> wrote in message news:<Xns952D6E17BC952jyanikkuanet@204.117.192.21>...

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407202154.764c02bf@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns952B71ED9A153jyanikkuanet@204.117.192.21>...
and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407190004.a0b4d86@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...
john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

snipp..
Be careful not to get more than 49 v DC in this step (or the
crowbar circuit will trigger).

If you have excessive current, check also the flyback diode in the
primary SMPS circuit. It's likely to blow, and causes a nasty short
circuit for the 1000 uF cap. A 200V 2 A fast switch ( < 80
nanosec) diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be
replaced with a TL 494 ( the 594 is a little bit harder to get, at
least here on the european continent).

hth,
andreas


The crowbar circuit is what was fried on his unit,I believe.
When the FET shorted,the full line V was applied to the crowbar.

I surmise that Q908 is the SCR,CR908 is the zener,and R909 is the 3
ohm fusible ww reistor.
And yes,the TL594/494 could have blown along with the FET.

Hi Jim,

well, you're right: In most of the 2235 I repaired with a fried smps
the fast switch diode is OK. I only had it once in a case of 15
broken 2235's I had to repair that the fast switch diode was blown ( a
dead short diode). This was a sour job: It took a while to diagnose
this, but finally turned out.

Sympthoms were: blown fuse, dead IC, dead FET and dead fast switch.
Crowbar was intact and OK.

Wel, but after rethinking this: The problem can be more clever
diagnosed when a variac is handy: With a good fuse (and diagnosed GOOD
crowbar) and the dead short FET in the scope, scope should power up
partly when one raises AC inlet voltage from 0 to 30V AC. If it powers
up, the fast switch diode is ok with confidence 99.9%. If not: I'd
recommend you check the diode closely.

Yes, one could use a curve tracer too for powering the scope - if
you're lucky to have one ;-)

hth,
Andreas


I don't think the pre-reg will fire up at such a low line V. There's a
couple of xstrs that supply power to the TL594 pin 12 that act as a delay
to allow the input line filter cap to charge up before the PWM IC starts.It
will also act as an undervoltage lockout.(normal op range is 90-250 VAC)
IIRC,line VAC has to get up to 60 VAC or greater for the IC to start.
One can hook up a LV ps across the "housekeeping cap" for the TL594 and
slowly raise the V to ~16 VDC and the IC will start,and a pulse train can
be scopes at pin10 or a ramp at pin5.

With the scope NOT connected to mains!!!

hi Jim,

the pre-reg has not (and should not) to fire up in the test I
described: When the FET is shorted, the the pre-req is not working.
The shorted pre-reg FET is a necessity for my test.

What I wanted to do is to isolate the fault:
a. Is ist the fast switch diode?
or
b: is it the crowbar.

hope this make more clear what the proposed test is for.

hth,
Andreas

 

[Jim Yanik]

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407222229.49a00e53@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns952D6E17BC952jyanikkuanet@204.117.192.21>...
and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407202154.764c02bf@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns952B71ED9A153jyanikkuanet@204.117.192.21>...
and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407190004.a0b4d86@posting.google.com:

Jim Yanik <jyanik@abuse.gov> wrote in message
news:<Xns9528C97961AC3jyanikkuanet@204.117.192.21>...
john <john@removeagent-j.thiscom> wrote in
newspsa8wpymt64p9b0@funkmeister.agent-j.com:

snipp..
Be careful not to get more than 49 v DC in this step (or the
crowbar circuit will trigger).

If you have excessive current, check also the flyback diode in
the primary SMPS circuit. It's likely to blow, and causes a
nasty short circuit for the 1000 uF cap. A 200V 2 A fast switch
( < 80 nanosec) diode will be a good replacement.

If the TL594 SMPS IC in the primary SMPS is blown, it can be
replaced with a TL 494 ( the 594 is a little bit harder to get,
at least here on the european continent).

hth,
andreas


The crowbar circuit is what was fried on his unit,I believe.
When the FET shorted,the full line V was applied to the crowbar.

I surmise that Q908 is the SCR,CR908 is the zener,and R909 is the
3 ohm fusible ww reistor.
And yes,the TL594/494 could have blown along with the FET.

Hi Jim,

well, you're right: In most of the 2235 I repaired with a fried
smps the fast switch diode is OK. I only had it once in a case of
15 broken 2235's I had to repair that the fast switch diode was
blown ( a dead short diode). This was a sour job: It took a while
to diagnose this, but finally turned out.

Sympthoms were: blown fuse, dead IC, dead FET and dead fast switch.
Crowbar was intact and OK.

Wel, but after rethinking this: The problem can be more clever
diagnosed when a variac is handy: With a good fuse (and diagnosed
GOOD crowbar) and the dead short FET in the scope, scope should
power up partly when one raises AC inlet voltage from 0 to 30V AC.
If it powers up, the fast switch diode is ok with confidence 99.9%.
If not: I'd recommend you check the diode closely.

Yes, one could use a curve tracer too for powering the scope - if
you're lucky to have one ;-)

hth,
Andreas


I don't think the pre-reg will fire up at such a low line V. There's
a couple of xstrs that supply power to the TL594 pin 12 that act as a
delay to allow the input line filter cap to charge up before the PWM
IC starts.It will also act as an undervoltage lockout.(normal op
range is 90-250 VAC) IIRC,line VAC has to get up to 60 VAC or greater
for the IC to start. One can hook up a LV ps across the "housekeeping
cap" for the TL594 and slowly raise the V to ~16 VDC and the IC will
start,and a pulse train can be scopes at pin10 or a ramp at pin5.

With the scope NOT connected to mains!!!

hi Jim,

the pre-reg has not (and should not) to fire up in the test I
described: When the FET is shorted, the the pre-req is not working.
The shorted pre-reg FET is a necessity for my test.

What I wanted to do is to isolate the fault:
a. Is ist the fast switch diode?
or
b: is it the crowbar.

hope this make more clear what the proposed test is for.

hth,
Andreas

You saying "scope should power up" had me thinking otherwise.
Now I'm NOT clear as to what's supposed to "power up" in your test.

Do you mean the rectified line V will climb if the diode is not shorted,and
the fuse will blow if the diode IS shorted?

You can just disconnect one end of the diode and eliminate it as a fault.
The TL594 can be tested without a FET in place to see if it runs,easiest
with a separate LV PS as I said.

--
Jim Yanik
jyanik-at-kua.net

 

[TekMan]

Jim Yanik <jyanik@abuse.gov> wrote in message news:<Xns952F6B4AB146Ejyanikkuanet@204.117.192.21>...

and7@bigfoot.com (TekMan) wrote in
news:6a624601.0407222229.49a00e53@posting.google.com:
...snippety for convenience..


I don't think the pre-reg will fire up at such a low line V. There's
a couple of xstrs that supply power to the TL594 pin 12 that act as a
delay to allow the input line filter cap to charge up before the PWM
IC starts.It will also act as an undervoltage lockout.(normal op
range is 90-250 VAC) IIRC,line VAC has to get up to 60 VAC or greater
for the IC to start. One can hook up a LV ps across the "housekeeping
cap" for the TL594 and slowly raise the V to ~16 VDC and the IC will
start,and a pulse train can be scopes at pin10 or a ramp at pin5.

With the scope NOT connected to mains!!!

hi Jim,

the pre-reg has not (and should not) to fire up in the test I
described: When the FET is shorted, the the pre-req is not working.
The shorted pre-reg FET is a necessity for my test.

What I wanted to do is to isolate the fault:
a. Is ist the fast switch diode?
or
b: is it the crowbar.

hope this make more clear what the proposed test is for.

hth,
Andreas


You saying "scope should power up" had me thinking otherwise.
Now I'm NOT clear as to what's supposed to "power up" in your test.

Do you mean the rectified line V will climb if the diode is not shorted,and
the fuse will blow if the diode IS shorted?

You can just disconnect one end of the diode and eliminate it as a fault.
The TL594 can be tested without a FET in place to see if it runs,easiest
with a separate LV PS as I said.

hi Jim,

sorry for my poor english: I try to say with "power up": apply power
and make the scope show some reaction.

Exactly as you mentioned: Let's assume a shorted FET in the front-end
pre-req, and a shorted fast switch diode. If you raise the voltage
with the variac, the current will raise quickly and the fuse blows.

If you have a shorted FET, and an good fast switch diode, the voltage
will raise. At a certain point, (approx. 35 to 40 V rectified DC
accross the 1000 uf Cap). the inverter ( located after the pre-req)
will start.

Of course you can test the diode by lifting one end, but the idea i
had (doing the test as described): You get a more easily diagnose of
the fault without opening the case before you have to open the case.

Of course everybody has it's own ways to start fault search. So don't
mind, I just wanted to give my idea. To share with the newsgroup..

hth,
Andreas

 

[pindiode]

hi
i have a 2235 (mil. version)
have the same problem with p/s
q908 was removed, does anyone have a part#
for this item . and possibly a scan of
service manual for p/s
thanks doug