info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Fender Frontman 212R Volume 80% down problem
934 views
Skip to first unread message
capossio...@gmail.com
Oct 14, 2015, 11:18:15 AM10/14/15
to
Hello,
I was wondering if you could help repair a guitar amplifier, it is a
Fender Frontman 212R, schematics are here:
http://support.fender.com/schematics/guitar_amplifiers/FM212R_schematic.pdf
The problems are:
-Low volume. It is a 100W amplifier and at full power is delivering
something like 20W tops at max volume.
-Paper crackling sound when guitar is played very hard. It is very
subtle, but is annoying.
I know that:
-Power supply voltages are ok (including the power amp stage)
-Speakers are ok (it has two speakers of 8 ohms in parallel as load)
-Nothing seems to be burnt on the PCB, or any cap seems bloated or something
It is unknown to me what the PTC (RT2) is doing, I know the NTC (RT1)
is for mains thermal shutdown, but the PTC seems to be for thermal
shutdown of the power stage only (because it is thermally coupled to
the power transistors power dissipator with some paste).
The next time I see it I will try to see if it is a pre-amp problem or
a power amp problem (because fortunately it has a pre-amp out and pwr
amp in)
Thanks.
I was wondering if you could help repair a guitar amplifier, it is a
Fender Frontman 212R, schematics are here:
http://support.fender.com/schematics/guitar_amplifiers/FM212R_schematic.pdf
The problems are:
-Low volume. It is a 100W amplifier and at full power is delivering
something like 20W tops at max volume.
-Paper crackling sound when guitar is played very hard. It is very
subtle, but is annoying.
I know that:
-Power supply voltages are ok (including the power amp stage)
-Speakers are ok (it has two speakers of 8 ohms in parallel as load)
-Nothing seems to be burnt on the PCB, or any cap seems bloated or something
It is unknown to me what the PTC (RT2) is doing, I know the NTC (RT1)
is for mains thermal shutdown, but the PTC seems to be for thermal
shutdown of the power stage only (because it is thermally coupled to
the power transistors power dissipator with some paste).
The next time I see it I will try to see if it is a pre-amp problem or
a power amp problem (because fortunately it has a pre-amp out and pwr
amp in)
Thanks.
N_Cook
Oct 14, 2015, 11:28:01 AM10/14/15
to
to narrow to PA or PREA.
Suddenly happen or previous intermittant level change?
Leonardo Capossio
Oct 14, 2015, 11:51:44 AM10/14/15
to
The crackling sound seems to appear when strongly strumming the guitar. The volume does not go up or down, it just never reaches a level above lets say 20W.
Ian Field
Oct 14, 2015, 1:09:49 PM10/14/15
to
"Leonardo Capossio" <capossio...@gmail.com> wrote in message
news:7cdfcf24-c1a0-4479...@googlegroups.com...
making it intermittently on big peaks. RoHS solder can cause things like
that.
It it turns out to be the low level stages, look for any discrete transistor
(or JFET) stages, a high ESR emitter bypass cap can allow a lot of
unintended nfb in that stage, so its gain will be a lot less than designed
for.
Any electrolytics coupling one stage to the next are also a possibility - if
you find any tantalum beads, study the circuit and satisfy yourself that
they can't under any circumstances be subjected to reverse voltage, they can
go leaky at the drop of a hat! If you find any tantalum caps in signal
paths, its not a bad idea to replace them with multilayer ceramic chip
types - some manufacturers offer them up to 180uF, and a few can supply
resin dipped leaded types.
Generally speaking; start with the soldering, then check electrolytics
(especially small ones). Resistors can go high in value or open circuit -
high resistance parts are particularly prone to increasing, the lower values
(especially wire wound) tend to fail open circuit, these are also slightly
more likely to show a burn mark.
If you find any scorched or burned open resistors, there's a good case to
start checking for leaky or shorted semiconductors.
Gareth Magennis
Oct 14, 2015, 2:42:38 PM10/14/15
to
wrote in message
news:b8a6f8b3-174b-47a6...@googlegroups.com...
switch gets dirty it can cause your symptoms.
If you take the main board out you will find a couple of discoloured hot
spots, particularly around the zener diodes providing the +- 15v for the
pre-amp.
These need sorting out before they become dry joints, if they haven't
already.
I had one of these not so long ago that had dry joints causing one half of
the push pull circuit not working, as Ian suggests.
The best solution here is to resolder EVERY joint on the mainboard. That
might sound like overkill, but I bet it would take you less than 10
minutes - probably less time than in would take to track down the single (at
present) dry joint.
Gareth.
Leonardo Capossio
Oct 14, 2015, 4:30:00 PM10/14/15
to
Taking out the PCB revealed that three 5W 0.22 ohm resistors were not original, but had been changed by someone (whom the owner cannot pinpoint), and this someone changed them for 22ohm resistors (one .22 ohm from the original remained, and this is probably causing the imbalance). I will replace this resistors with correct ones, and see if this fixes all the problems.
Ian Field
Oct 14, 2015, 4:57:51 PM10/14/15
to
news:c922cd09-fa98-4ccd...@googlegroups.com...
Those parts had probably been replaced because the originals blew. Check
whether any associated semiconductors show any sign of having been replaced
as well - it wouldn't hurt to test them anyway.
whether any associated semiconductors show any sign of having been replaced
as well - it wouldn't hurt to test them anyway.
jurb...@gmail.com
Oct 14, 2015, 9:16:32 PM10/14/15
to
How are you measuring this power output to be 20 watts ?
And as for what is happening here, it is EXTREMELY hard to describe sounds with words, like second harmonic distortion n shit and how it differs from third harmonic.
Know what ? I got an idea and I can do it. Put up a website with different sounds describing what they are. People do not even know the difference between a 60 cycle hum and a 120 cycle hum. I can hear the effect of an open polarity in an amp hands down and go immediately to the problem which usually was a driver (not output) transistor with an open B-E or B-C junction. I could ID the problem by the sound.
Maybe what I should do is make these sounds and put them up on the net. I know how to disconnect this or that and make it sound like that exact fault.
^Yeah right. Bob Carver got his claim to fame mimicking the sound of super audiophile amps, well I am Cletus Carver and I am gong to have my claim to fame mimicking fucked up amps...
LOL LOL
And as for what is happening here, it is EXTREMELY hard to describe sounds with words, like second harmonic distortion n shit and how it differs from third harmonic.
Know what ? I got an idea and I can do it. Put up a website with different sounds describing what they are. People do not even know the difference between a 60 cycle hum and a 120 cycle hum. I can hear the effect of an open polarity in an amp hands down and go immediately to the problem which usually was a driver (not output) transistor with an open B-E or B-C junction. I could ID the problem by the sound.
Maybe what I should do is make these sounds and put them up on the net. I know how to disconnect this or that and make it sound like that exact fault.
^Yeah right. Bob Carver got his claim to fame mimicking the sound of super audiophile amps, well I am Cletus Carver and I am gong to have my claim to fame mimicking fucked up amps...
LOL LOL
Phil Allison
Oct 15, 2015, 1:01:42 AM10/15/15
to
Leonardo Capossio wrote:
>
>
> Taking out the PCB revealed that three 5W 0.22 ohm resistors were not
> original, but had been changed by someone (whom the owner cannot pinpoint)
>
>
> Taking out the PCB revealed that three 5W 0.22 ohm resistors were not
> original, but had been changed by someone (whom the owner cannot pinpoint)
> and this someone changed them for 22ohm resistors (one .22 ohm from the
> original remained, and this is probably causing the imbalance). I will
> replace this resistors with correct ones, and see if this fixes all
> the problems.
** That is bad error to make, the repairer must have been clueless or blind.
> original remained, and this is probably causing the imbalance). I will
> replace this resistors with correct ones, and see if this fixes all
> the problems.
I see four 0.22ohm, 5W resistors on the schem - two in the power amp and two ( wired in parallel) in series with the ground link for the speaker.
The only time you need to change such resistors is when output transistors have failed and caused them to burn out - so make sure they are the correct type numbers and mounted properly to the heatsink.
... Phil
Leonardo Capossio
Oct 15, 2015, 5:11:29 PM10/15/15
to
Well, in the end the problem were those resistors, now it works like it should. The only problem is that at high volumes the amp turns the sound off after a while, I guess because of thermal shutdown of the PT100 on the power dissipator. But this can be solved by cleaning and re-applying thermal paste, and in extreme situation adding a cooler.
Phil Allison
Oct 16, 2015, 12:54:16 AM10/16/15
to
Often there are burrs left around bolt holes in the coupler that prevent the metal surfaces mating properly.
Those holes need to be de-burred in a drill press.
Once done and regreased, output devices run cooler and there are no more shut downs.
... Phil
Ian Field
Oct 16, 2015, 12:52:04 PM10/16/15
to
"Phil Allison" <palli...@gmail.com> wrote in message
news:657c1a61-421d-4e07...@googlegroups.com...
> Leonardo Capossio wrote:
>>
>>
>> Well, in the end the problem were those resistors,
>> now it works like it should. The only problem is that
>> at high volumes the amp turns the sound off after a while,
>> I guess because of thermal shutdown of the PT100 on the
>> power dissipator. But this can be solved by cleaning and
>> re-applying thermal paste, and in extreme situation
>> adding a cooler.
>
> ** That Fender model and similar ones uses a thick, alloy bar to couple
> heat from the output devices to the bottom of the case - right ?
>
> Often there are burrs left around bolt holes in the coupler that prevent
> the metal surfaces mating properly.
>
> Those holes need to be de-burred in a drill press.
If the parts are malleable metal, a drill press can be difficult to control
>>
>>
>> Well, in the end the problem were those resistors,
>> now it works like it should. The only problem is that
>> at high volumes the amp turns the sound off after a while,
>> I guess because of thermal shutdown of the PT100 on the
>> power dissipator. But this can be solved by cleaning and
>> re-applying thermal paste, and in extreme situation
>> adding a cooler.
>
> ** That Fender model and similar ones uses a thick, alloy bar to couple
> heat from the output devices to the bottom of the case - right ?
>
> Often there are burrs left around bolt holes in the coupler that prevent
> the metal surfaces mating properly.
>
> Those holes need to be de-burred in a drill press.
how much metal is removed.
Usually a 3/8" (10mm) drill bit held in the hand is quite sufficient.
Jeroni Paul
Oct 16, 2015, 2:53:56 PM10/16/15
to
I recall a similar case where I found a big heavy 25W resistor fitted in the line level audio path of a JAMMA adapter. When I found the schematic I understood, it.called for a .25W resistor but the decimal point was barely visible.
M Philbrook
Oct 16, 2015, 7:59:00 PM10/16/15
to
In article <NW9Ux.42147$rO5....@fx39.am4>,
gangprob...@ntlworld.com says...
A center drill,tapered counterbore/sink or chamfer cutters are great for
task like that.
Make sure its metal grade..
Jamie
gangprob...@ntlworld.com says...
task like that.
Make sure its metal grade..
Jamie
Phil Allison
Oct 17, 2015, 1:55:50 AM10/17/15
to
Ian Field wrote:
>
> "Phil Allison"
The cast alloy was a bit too hard for a drill bit held in the fingers.
... Phil
>
> "Phil Allison"
>
> >
> > ** That Fender model and similar ones uses a thick, alloy bar to couple
> > heat from the output devices to the bottom of the case - right ?
> >
> > Often there are burrs left around bolt holes in the coupler that prevent
> > the metal surfaces mating properly.
> >
> > Those holes need to be de-burred in a drill press.
>
>
> If the parts are malleable metal, a drill press can be difficult to control
> how much metal is removed.
** Done about a dozen of them that way, it was quick & easy with a light touch on the feed handle. There are 5 or 6 holes that need doing at both ends.
> >
> > ** That Fender model and similar ones uses a thick, alloy bar to couple
> > heat from the output devices to the bottom of the case - right ?
> >
> > Often there are burrs left around bolt holes in the coupler that prevent
> > the metal surfaces mating properly.
> >
> > Those holes need to be de-burred in a drill press.
>
>
> If the parts are malleable metal, a drill press can be difficult to control
> how much metal is removed.
The cast alloy was a bit too hard for a drill bit held in the fingers.
... Phil
Ian Field
Oct 18, 2015, 11:39:57 AM10/18/15
to
news:0b2eab1a-cf45-4175...@googlegroups.com...
Usually I find cast alloy heatsinks in automotive assemblies and power
tools, they seem to be the exception in most other things.
The common rolled sheet aluminium is very easy to cut too deep, I think
extruded is a little harder, but it still needs a light touch if you do it
in a drill press.
Anything other than a block can also snag on the drill bit if you
accidentally go right through.
tools, they seem to be the exception in most other things.
The common rolled sheet aluminium is very easy to cut too deep, I think
extruded is a little harder, but it still needs a light touch if you do it
in a drill press.
Anything other than a block can also snag on the drill bit if you
accidentally go right through.
Leonardo Capossio
Nov 24, 2015, 4:37:24 PM11/24/15
to
El jueves, 15 de octubre de 2015, 18:11:29 (UTC-3), Leonardo Capossio escribió:
> Well, in the end the problem were those resistors, now it works like it should. The only problem is that at high volumes the amp turns the sound off after a while, I guess because of thermal shutdown of the PT100 on the power dissipator. But this can be solved by cleaning and re-applying thermal paste, and in extreme situation adding a cooler.
Unfortunately after re-applying thermal paste on the heatsink and the power transistors, and adding a cooler on the far side (not exactly by the heatsink) to blow hot air out of the amp head, the amp still behaves the same. It turns off the audio (LED and cooler are still working) after a couple of minutes of playing.
> Well, in the end the problem were those resistors, now it works like it should. The only problem is that at high volumes the amp turns the sound off after a while, I guess because of thermal shutdown of the PT100 on the power dissipator. But this can be solved by cleaning and re-applying thermal paste, and in extreme situation adding a cooler.
Anyone can point out to what might be the problem ?
M Philbrook
Nov 24, 2015, 5:58:27 PM11/24/15
to
In article <b9e7a5a8-3dd2-4875...@googlegroups.com>,
capossio...@gmail.com says...
Does the outputs on the sink actually feel like they are getting that
hot?
DO you have a IR tool to meaure it? Most likely not..
If you feel the unit isn't that hot and should be able to maintain
output at that temp? then maybe the PT resistor has degraded. They do
drift a little over time.
You could try checking all of the R's in that ciccuit to make sure they
are on track.
Looking at the schematic it looks like you have at least 2 R's that
could be suspect. That would be in the lower left about 1/2 the way
over.
If you have some kind of component cooler you can spray the PT while
its operating and see if it extends the on time. Or blow on it using a
straw.
Jamie
capossio...@gmail.com says...
hot?
DO you have a IR tool to meaure it? Most likely not..
If you feel the unit isn't that hot and should be able to maintain
output at that temp? then maybe the PT resistor has degraded. They do
drift a little over time.
You could try checking all of the R's in that ciccuit to make sure they
are on track.
Looking at the schematic it looks like you have at least 2 R's that
could be suspect. That would be in the lower left about 1/2 the way
over.
If you have some kind of component cooler you can spray the PT while
its operating and see if it extends the on time. Or blow on it using a
straw.
Jamie
dansabr...@yahoo.com
Nov 24, 2015, 6:10:30 PM11/24/15
to
Put a scope on the outputs and look for very high frequency harmonics. Perhaps the system is working too hard on things you can not hear.
Dan
Dan
mako...@yahoo.com
Nov 25, 2015, 10:26:48 AM11/25/15
to
On Tuesday, November 24, 2015 at 6:10:30 PM UTC-5, dansabr...@yahoo.com wrote:
> Put a scope on the outputs and look for very high frequency harmonics. Perhaps the system is working too hard on things you can not hear.
>
> Dan
good point
> Put a scope on the outputs and look for very high frequency harmonics. Perhaps the system is working too hard on things you can not hear.
>
> Dan
logical seteps
is it actually getting hot?
if not, the thermal shutdown system could be faulty
if it is getting hot you need to figure out why....
there could be an ultrasonic oscilation in the amp
the quiescent bias current could be too high in the amp
there could be some kind of short or overload at the output of the amp, bad speaker?
have fun , fixing stuff is a great way to learn
Mark
Leonardo Capossio
Nov 25, 2015, 11:13:52 AM11/25/15
to
Just to clariphy, the audio from the amp gets cut abruptly. Everything is working like a charm, the sound isn't fading away or anything, and the suddenly it stops outputting audio. Everything else seems to be working.
In this case, the only thermally dependent element I see is the PTC100. The circuit surrounding the PT100 seems to be compensating the bias current of the power amp, though I do not understand it fully, if anyone can point to a source that explains how this circuit work it would be great. To me it seems very strange that it abruptly cuts out the sound.
dansabr...@yahoo.com
Nov 25, 2015, 12:24:18 PM11/25/15
to
If there are ultrasonic frequencies being amplified, you wont hear them. But... they will cause the amp to work really hard and generate more heat than usual. This could cause the amp to shutdown due to the heat. Other than using a scope, there is no way to confirm this.
Dan
Dan
Leonardo Capossio
Nov 25, 2015, 12:48:43 PM11/25/15
to
Leonardo Capossio
Nov 25, 2015, 12:50:58 PM11/25/15
to
El miércoles, 25 de noviembre de 2015, 14:24:18 (UTC-3), dansabr...@yahoo.com escribió:
> If there are ultrasonic frequencies being amplified, you wont hear them. But... they will cause the amp to work really hard and generate more heat than usual. This could cause the amp to shutdown due to the heat. Other than using a scope, there is no way to confirm this.
>
> Dan
If there are ultrasonic frequencies (which I cannot measure), then what could cause this ? Where might they be generated ?
> If there are ultrasonic frequencies being amplified, you wont hear them. But... they will cause the amp to work really hard and generate more heat than usual. This could cause the amp to shutdown due to the heat. Other than using a scope, there is no way to confirm this.
>
> Dan
The amp is shutting down via the PT100 circuit, that should also do some temperature drift compensation. If I bypass this compensation, would the amp sound very bad ?
jerry...@gmail.com
Nov 25, 2015, 4:18:24 PM11/25/15
to
Leonardo Capossio
Nov 25, 2015, 9:18:10 PM11/25/15
to
Phil Allison
Nov 25, 2015, 10:23:16 PM11/25/15
to
Leonardo Capossio wrote:
>
>
>
> In this case, the only thermally dependent element I see is the PTC100.
> The circuit surrounding the PT100 seems to be compensating the bias
> current of the power amp, though I do not understand it fully, if
> anyone can point to a source that explains how this circuit work it
> would be great.
** PTC100 operates the "mute" function of the power amp by shutting off Q7 and so the base voltage to Q8 - which is a 2mA current source for the input pair Q9 & Q10. This results in no current in the driver stage - transistors Q11, Q12 & Q13 and so no amplification.
>
>
>
> In this case, the only thermally dependent element I see is the PTC100.
> The circuit surrounding the PT100 seems to be compensating the bias
> current of the power amp, though I do not understand it fully, if
> anyone can point to a source that explains how this circuit work it
> would be great.
.... Phil
Leonardo Capossio
Nov 26, 2015, 12:04:28 PM11/26/15
to
The mute coming from the input jacks seems to pull the "mute" point to ground, hence setting the base voltage much lower than 10v+0.7v, and thus shutting down Q7.
Now I don't fully understand how the PTC100 shuts it down. At 25deg the PTC100 has 110ohm, according to some tables, and at 100deg it has a little less than 140ohms. I don't know how to calculate Q6's Q point (Vec or Ie), since Veb/PTC100 seems to control Ie directly (unless the only way out is using Shockley's equation). Also the shut down seems way to abrupt, like Q6 is open, and then suddenly it is a closed switch.
mako...@yahoo.com
Nov 26, 2015, 11:50:10 PM11/26/15
to
Is the heat sink hot when the unit shuts down?
Use your hand, if you can't keep your hand on the
Heat sink, its too hot.
Use your hand, if you can't keep your hand on the
Heat sink, its too hot.
Phil Allison
Nov 27, 2015, 1:05:29 AM11/27/15
to
Leonardo Capossio wrote:
>
>>
> > ** PTC100 operates the "mute" function of the power amp by shutting off Q7 and so the base voltage to Q8 - which is a 2mA current source for the input pair Q9 & Q10. This results in no current in the driver stage - transistors Q11, Q12 & Q13 and so no amplification.
> >
> >
>
>
>>
> > ** PTC100 operates the "mute" function of the power amp by shutting off Q7 and so the base voltage to Q8 - which is a 2mA current source for the input pair Q9 & Q10. This results in no current in the driver stage - transistors Q11, Q12 & Q13 and so no amplification.
> >
> >
>
> Now I don't fully understand how the PTC100 shuts it down. At 25deg the PTC100 has 110ohm, according to some tables, and at 100deg it has a little less than 140ohms.
** There is a device called PT100, a platinum resistance sensor that is quite linear and rises about 0.4% for each degree C. The thing on the schem labelled "PTC100" is not one of them.
It is a PTC "positive temp coefficient" thermistor (aka posistor) with a steep rise in resistance at 100C. The mute circuit in the Fender trips when the PTC device reaches 800 ohms and is rising steeply - so operates with a snap action.
FYI: "PTC100" is not a part number, but a code used on the schem referring to the parts list.
... Phil
Leonardo Capossio
Nov 27, 2015, 4:13:56 PM11/27/15
to
Leonardo Capossio
Nov 27, 2015, 4:45:55 PM11/27/15
to
Phil Allison
Nov 27, 2015, 10:43:39 PM11/27/15
to
Leonardo Capossio wrote:
>
>
>
> > > Now I don't fully understand how the PTC100 shuts it down.
> > > At 25deg the PTC100 has 110ohm, according to some tables,
> > > and at 100deg it has a little less than 140ohms.
> >
> >
> > ** There is a device called PT100, a platinum resistance sensor
> > that is quite linear and rises about 0.4% for each degree C. The
> > thing on the schem labelled "PTC100" is not one of them.
> >
> > It is a PTC "positive temp coefficient" thermistor (aka posistor)
> > with a steep rise in resistance at 100C. The mute circuit in the
> > Fender trips when the PTC device reaches 800 ohms and is rising
> > steeply - so operates with a snap action.
> >
> > FYI: "PTC100" is not a part number, but a code used on the schem
> > referring to the parts list.
> >
> >
>
>
>
>
> > > Now I don't fully understand how the PTC100 shuts it down.
> > > At 25deg the PTC100 has 110ohm, according to some tables,
> > > and at 100deg it has a little less than 140ohms.
> >
> >
> > ** There is a device called PT100, a platinum resistance sensor
> > that is quite linear and rises about 0.4% for each degree C. The
> > thing on the schem labelled "PTC100" is not one of them.
> >
> > It is a PTC "positive temp coefficient" thermistor (aka posistor)
> > with a steep rise in resistance at 100C. The mute circuit in the
> > Fender trips when the PTC device reaches 800 ohms and is rising
> > steeply - so operates with a snap action.
> >
> > FYI: "PTC100" is not a part number, but a code used on the schem
> > referring to the parts list.
> >
> >
>
> That explains a LOT. Thanks.
>
>
** BTW, the NTC10-15 shown in series with the AC power is used as an anti-surge device. It is mounted on the PCB, not inside the transformer.
>
>
At room temp it's a 10ohm resistor, falling to 0.2ohms at 5 amps current.
At switch-on it simply adds 10 ohms to the wiring on the primary side, limiting inrush surge currents so the fuse does not blow regularly.
.... Phil
Leonardo Capossio
Nov 29, 2015, 1:47:06 PM11/29/15
to
I'm starting to think this is the problem. I checked the trasistors and they are fine, they are the same as in the schematic.
Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors, between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot as hell (can't even touch it) and the heatsink is very hot, but not so much.
I wonder if there is a video on how to do it properly.
Ralph Mowery
Nov 29, 2015, 2:15:36 PM11/29/15
to
"Leonardo Capossio" <capossio...@gmail.com> wrote in message
news:14fca79b-407a-43dc...@googlegroups.com...
>Y removed them from the heatsink, cleaned everything with alcohol (even
>tried not to touch it with my fingers) and put fresh thermal paste (with
>isolation) between the metal plate that holds the transitors, between the
>plate and the heatsink, and between the heatsink and the case, but still
>the metal plate is hot ?>as hell (can't even touch it) and the heatsink is
>tried not to touch it with my fingers) and put fresh thermal paste (with
>isolation) between the metal plate that holds the transitors, between the
>plate and the heatsink, and between the heatsink and the case, but still
>very hot, but not so much.
When putting the paste on, did you use the thinist layer you could ? The
paste is not that conductive to heat, just beter than air. The ideal
ammount (which probably can not be done) would be to have the metel surfaces
touching each other on the high points and just the voids filled with the
paste.
Less is often beter than more.
Dave Platt
Nov 29, 2015, 3:07:31 PM11/29/15
to
In article <14fca79b-407a-43dc...@googlegroups.com>,
If the plate was warped at all, it won't be making good contact with
the heatsink - there would be a few small points or lines of good
contact, and a significant air-gap everywhere else. Filling the gap
with thermal paste won't help very much. Thermal paste helps thermal
conductivity only when it's a very thin layer - just thick enough to
bridge the occasional thin gap between two surfaces which are
otherwise in good direct physical contact. A thick layer may be worse
than not having it at all.
CPU overclockers used to (maybe still) "lap" the top of the CPU and
the contact surface of the heatsink, in order to get the both as flat
as possible. When this is done, the layer of thermal paste between
the two is *very* thin... put a tiny dab on one surface, and then
spread it out thin-and-even with a single-edged razor blade.
Over-torquing the screws which hold down TO-220 or similar
plastic-package transistors to the heatsink can similarly cause
problems. Apply too much torque, make the transistor case and tab
flex a bit, and you end up pulling most of the case away from the
heatsink and thermal transfer gets worse. If your transistor tabs are
bent or warped, lapping them might help.
[Lapping of this type can be done with a sheet of wet-and-dry carbide
paper, mounted on a flat plate of some sort - a high-quality sheet of
plate glass will serve. The usual finishing technique of "start with
a coarser grit, then work your way up through the grits until you get
to 1000" would be appropriate.]
The fact that the heatsink itself is very hot does suggest that
something else may be going amiss... maybe the transistor quiescent
bias is too high, or maybe there's a parasitic oscillation taking
place. Using a scope (or an old-style grid-dip meter in absorbtion
mode) to look for RF where it doesn't belong might be a good idea.
Leonardo Capossio <capossio...@gmail.com> wrote:
>Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my
>fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors,
>between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot
>as hell (can't even touch it) and the heatsink is very hot, but not so much.
You might want to check the metal plate and heatsink for flatness.
>Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my
>fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors,
>between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot
>as hell (can't even touch it) and the heatsink is very hot, but not so much.
If the plate was warped at all, it won't be making good contact with
the heatsink - there would be a few small points or lines of good
contact, and a significant air-gap everywhere else. Filling the gap
with thermal paste won't help very much. Thermal paste helps thermal
conductivity only when it's a very thin layer - just thick enough to
bridge the occasional thin gap between two surfaces which are
otherwise in good direct physical contact. A thick layer may be worse
than not having it at all.
CPU overclockers used to (maybe still) "lap" the top of the CPU and
the contact surface of the heatsink, in order to get the both as flat
as possible. When this is done, the layer of thermal paste between
the two is *very* thin... put a tiny dab on one surface, and then
spread it out thin-and-even with a single-edged razor blade.
Over-torquing the screws which hold down TO-220 or similar
plastic-package transistors to the heatsink can similarly cause
problems. Apply too much torque, make the transistor case and tab
flex a bit, and you end up pulling most of the case away from the
heatsink and thermal transfer gets worse. If your transistor tabs are
bent or warped, lapping them might help.
[Lapping of this type can be done with a sheet of wet-and-dry carbide
paper, mounted on a flat plate of some sort - a high-quality sheet of
plate glass will serve. The usual finishing technique of "start with
a coarser grit, then work your way up through the grits until you get
to 1000" would be appropriate.]
The fact that the heatsink itself is very hot does suggest that
something else may be going amiss... maybe the transistor quiescent
bias is too high, or maybe there's a parasitic oscillation taking
place. Using a scope (or an old-style grid-dip meter in absorbtion
mode) to look for RF where it doesn't belong might be a good idea.
M Philbrook
Nov 29, 2015, 9:00:42 PM11/29/15
to
In article <14fca79b-407a-43dc...@googlegroups.com>,
capossio...@gmail.com says...
If it's getting hot like that with the amp on and not generating any
sound, you have other problems.
Put a volt meter on the output in DC mode first, see if you are getting
any DC above a few mVs
After that, put the meter in AC mode and see if you are getting AC
ripple of a significant level, which you should not be.
Do this in ideal mode, no sound.
Me thinks you have big ISSUES with biasing.
Jamie.
capossio...@gmail.com says...
sound, you have other problems.
Put a volt meter on the output in DC mode first, see if you are getting
any DC above a few mVs
After that, put the meter in AC mode and see if you are getting AC
ripple of a significant level, which you should not be.
Do this in ideal mode, no sound.
Me thinks you have big ISSUES with biasing.
Jamie.
Phil Allison
Nov 29, 2015, 10:16:15 PM11/29/15
to
Leonardo Capossio wrote:
> >
> > ** That Fender model and similar ones uses a thick, alloy bar to couple heat from the output devices to the bottom of the case - right ?
> >
> > Often there are burrs left around bolt holes in the coupler that prevent the metal surfaces mating properly.
> >
> > Those holes need to be de-burred in a drill press.
> >
> > Once done and regreased, output devices run cooler and there are no more shut downs.
> >
> >
> > ... Phil
>
> Right.
>
> I'm starting to think this is the problem. I checked the trasistors and they are fine, they are the same as in the schematic.
>
> Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors, between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot as hell (can't even touch it) and the heatsink is very hot, but not so much.
>
> I wonder if there is a video on how to do it properly.
>
>
** Two things are not clear:
> >
> > ** That Fender model and similar ones uses a thick, alloy bar to couple heat from the output devices to the bottom of the case - right ?
> >
> > Often there are burrs left around bolt holes in the coupler that prevent the metal surfaces mating properly.
> >
> > Those holes need to be de-burred in a drill press.
> >
> > Once done and regreased, output devices run cooler and there are no more shut downs.
> >
> >
> > ... Phil
>
> Right.
>
> I'm starting to think this is the problem. I checked the trasistors and they are fine, they are the same as in the schematic.
>
> Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors, between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot as hell (can't even touch it) and the heatsink is very hot, but not so much.
>
> I wonder if there is a video on how to do it properly.
>
>
1. Does the heatsink get very hot whenever the amp is on or do you have to PLAY something ?
2. Have you found and fixed any burs on the holes in the heatsink coupler and or the chassis ?
BTW:
That amp has an odd, non adjustable bias arrangement.
Transistors Q14 and Q 15 should be in thermal contact with the heatsink.
... Phil
Leonardo Capossio
Nov 30, 2015, 9:23:54 AM11/30/15
to
El domingo, 29 de noviembre de 2015, 16:15:36 (UTC-3), Ralph Mowery wrote:
> When putting the paste on, did you use the thinist layer you could ? The
> paste is not that conductive to heat, just beter than air. The ideal
> ammount (which probably can not be done) would be to have the metel surfaces
> touching each other on the high points and just the voids filled with the
> paste.
> Less is often beter than more.
Yes, I used a very thin layer.
> When putting the paste on, did you use the thinist layer you could ? The
> paste is not that conductive to heat, just beter than air. The ideal
> ammount (which probably can not be done) would be to have the metel surfaces
> touching each other on the high points and just the voids filled with the
> paste.
> Less is often beter than more.
Leonardo Capossio
Nov 30, 2015, 9:29:36 AM11/30/15
to
On Monday, November 30, 2015 at 12:16:15 AM UTC-3, Phil Allison wrote:
> Leonardo Capossio wrote:
>
> > >
> > > ** That Fender model and similar ones uses a thick, alloy bar to couple heat from the output devices to the bottom of the case - right ?
> > >
> > > Often there are burrs left around bolt holes in the coupler that prevent the metal surfaces mating properly.
> > >
> > > Those holes need to be de-burred in a drill press.
> > >
> > > Once done and regreased, output devices run cooler and there are no more shut downs.
> > >
> > >
> > > ... Phil
> >
> > Right.
> >
> > I'm starting to think this is the problem. I checked the trasistors and they are fine, they are the same as in the schematic.
> >
> > Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors, between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot as hell (can't even touch it) and the heatsink is very hot, but not so much.
> >
> > I wonder if there is a video on how to do it properly.
> >
> >
>
> ** Two things are not clear:
>
> 1. Does the heatsink get very hot whenever the amp is on or do you have to PLAY something ?
>
You have to play for a while at low volumes (10min tops), or play for a minute at the highest volume. In idle, with a plug connected (otherwise the amp mutes), it does not shut down.
> Leonardo Capossio wrote:
>
> > >
> > > ** That Fender model and similar ones uses a thick, alloy bar to couple heat from the output devices to the bottom of the case - right ?
> > >
> > > Often there are burrs left around bolt holes in the coupler that prevent the metal surfaces mating properly.
> > >
> > > Those holes need to be de-burred in a drill press.
> > >
> > > Once done and regreased, output devices run cooler and there are no more shut downs.
> > >
> > >
> > > ... Phil
> >
> > Right.
> >
> > I'm starting to think this is the problem. I checked the trasistors and they are fine, they are the same as in the schematic.
> >
> > Y removed them from the heatsink, cleaned everything with alcohol (even tried not to touch it with my fingers) and put fresh thermal paste (with isolation) between the metal plate that holds the transitors, between the plate and the heatsink, and between the heatsink and the case, but still the metal plate is hot as hell (can't even touch it) and the heatsink is very hot, but not so much.
> >
> > I wonder if there is a video on how to do it properly.
> >
> >
>
> ** Two things are not clear:
>
> 1. Does the heatsink get very hot whenever the amp is on or do you have to PLAY something ?
>
> 2. Have you found and fixed any burs on the holes in the heatsink coupler and or the chassis ?
>
> BTW:
>
> That amp has an odd, non adjustable bias arrangement.
>
> Transistors Q14 and Q 15 should be in thermal contact with the heatsink.
>
>
Phil Allison
Dec 1, 2015, 12:05:16 AM12/1/15
to
Leonardo Capossio wrote:
>
> >
> > ** Two things are not clear:
> >
> > 1. Does the heatsink get very hot whenever the amp is on or do
> > you have to PLAY something ?
> >
>
> You have to play for a while at low volumes (10min tops), or play for a minute at the highest volume. In idle, with a plug connected (otherwise the amp mutes), it does not shut down.
** When in "mute" mode, the power stage is completely shut down and cannot get hot. But does the heatsink get hot when active but not played, even if this take an hour ?
>
> >
> > ** Two things are not clear:
> >
> > 1. Does the heatsink get very hot whenever the amp is on or do
> > you have to PLAY something ?
> >
>
> You have to play for a while at low volumes (10min tops), or play for a minute at the highest volume. In idle, with a plug connected (otherwise the amp mutes), it does not shut down.
> > 2. Have you found and fixed any burs on the holes in the heatsink
> > coupler and or the chassis ?
> >
>
> If by burr you mean what is described in this page: "http://www.nmri.go.jp/eng/khirata/metalwork/basic/bari/index_e.html", then I don't see any worrysome burr on the heatsink.
The best test is with a finger tip, the hole should not feel rough when you rub your finger over it.
>
> > BTW:
> >
> > That amp has an odd, non adjustable bias arrangement.
> >
> > Transistors Q14 and Q 15 should be in thermal contact with the heatsink.
> >
> >
>
> They are on the heatsink, but their tab is completely plastic.
What you are describing is called " thermal runaway " and I have seen it with other Fender models that use the same heatsink arrangement.
In those examples, careful de-burring of ALL the holes, re-greasing and making sure bias tracking devices were thermally coupled to the heatsink fixed the problem completely.
BTW:
You are lucky the Frontman 212 merely shuts down, the ones I worked on all blew up as there was no over temp sensing device fitted.
.... Phil
Leonardo Capossio
Dec 2, 2015, 1:20:59 PM12/2/15
to
On Tuesday, December 1, 2015 at 2:05:16 AM UTC-3, Phil Allison wrote:
> Leonardo Capossio wrote:
>
> >
> > >
> > > ** Two things are not clear:
> > >
> > > 1. Does the heatsink get very hot whenever the amp is on or do
> > > you have to PLAY something ?
> > >
> >
> > You have to play for a while at low volumes (10min tops), or play for a minute at the highest volume. In idle, with a plug connected (otherwise the amp mutes), it does not shut down.
>
> ** When in "mute" mode, the power stage is completely shut down and cannot get hot. But does the heatsink get hot when active but not played, even if this take an hour ?
>
>
After half an hour, it is mildly hot, don't think it gets worse over a longer period of time.
> Leonardo Capossio wrote:
>
> >
> > >
> > > ** Two things are not clear:
> > >
> > > 1. Does the heatsink get very hot whenever the amp is on or do
> > > you have to PLAY something ?
> > >
> >
> > You have to play for a while at low volumes (10min tops), or play for a minute at the highest volume. In idle, with a plug connected (otherwise the amp mutes), it does not shut down.
>
> ** When in "mute" mode, the power stage is completely shut down and cannot get hot. But does the heatsink get hot when active but not played, even if this take an hour ?
>
>
> > > 2. Have you found and fixed any burs on the holes in the heatsink
> > > coupler and or the chassis ?
> > >
> >
> > If by burr you mean what is described in this page: "http://www.nmri.go.jp/eng/khirata/metalwork/basic/bari/index_e.html", then I don't see any worrysome burr on the heatsink.
>
> ** Figure 7 is the appropriate one.
>
> The best test is with a finger tip, the hole should not feel rough when you rub your finger over it.
>
>
> >
> > > BTW:
> > >
> > > That amp has an odd, non adjustable bias arrangement.
> > >
> > > Transistors Q14 and Q 15 should be in thermal contact with the heatsink.
> > >
> > >
> >
> > They are on the heatsink, but their tab is completely plastic.
>
> ** That does not matter, but they do need to be held down and have a smear of thermal grease under them.
>
> What you are describing is called " thermal runaway " and I have seen it with other Fender models that use the same heatsink arrangement.
>
> In those examples, careful de-burring of ALL the holes, re-greasing and making sure bias tracking devices were thermally coupled to the heatsink fixed the problem completely.
>
> BTW:
>
> You are lucky the Frontman 212 merely shuts down, the ones I worked on all blew up as there was no over temp sensing device fitted.
>
>
> .... Phil
0 new messages