Hi!
Since the heatsinks of the old ripped Marantz amp I refitted with modern parts gets very hot I decided to also replace the old heatsink grease.
I unscrewed the amp from the sink, cleaned everything up, and placed about the same amount like the original grease of Arctic Silver III on the STK080G chips, then screwed everythin back together.
And now, o wonder, the heatsink gets even hotter!
After thinking a little bit about this I begin getting confused...
Since the heat transfer with the new grease from the ICs / transistors is better than with the 24 years old grease, is the heasink supposed to get hotter? Like, say, more heat is being transferred to the sink, so the sink gets hotter while the ICs stay cooler?
And how big is the influence a metal case on the heatsink? The sink originally was touching the metal case of the amp on three sides (sides and bottom), while nowadays it sits on wood... And I think that the heatsink never got so hot when the receiver was in its original state (with internal reveiver and pre-amp connected)...
You cannot really touch the heatsinks for very long - I have no way of measuring the temperature right now, but it's pretty hot. Should I maybe use a bigger heatsink, or are those STK chips not so picky about heating?
Mhmmm... maybe a fan would help... but I don't like that option...
Bye,
Arndt
Since the heatsinks of the old ripped Marantz amp I refitted with modern parts gets very hot I decided to also replace the old heatsink grease.
I unscrewed the amp from the sink, cleaned everything up, and placed about the same amount like the original grease of Arctic Silver III on the STK080G chips, then screwed everythin back together.
And now, o wonder, the heatsink gets even hotter!
After thinking a little bit about this I begin getting confused...
Since the heat transfer with the new grease from the ICs / transistors is better than with the 24 years old grease, is the heasink supposed to get hotter? Like, say, more heat is being transferred to the sink, so the sink gets hotter while the ICs stay cooler?
And how big is the influence a metal case on the heatsink? The sink originally was touching the metal case of the amp on three sides (sides and bottom), while nowadays it sits on wood... And I think that the heatsink never got so hot when the receiver was in its original state (with internal reveiver and pre-amp connected)...
You cannot really touch the heatsinks for very long - I have no way of measuring the temperature right now, but it's pretty hot. Should I maybe use a bigger heatsink, or are those STK chips not so picky about heating?
Mhmmm... maybe a fan would help... but I don't like that option...
Bye,
Arndt
Heat
Hi,
If you transfer more heat to the heatsink it gets hotter.
STK IC's normally don't get hot when on a heatsink except when running at full power.
I am sure the metal case did conduct much heat off the heatsink as the metal is a far more better conductor than air. Or see the metalcase as a extension of the heatsink. I am sure this is the cause for the hotter heatsink now.
Old Crown amplifier D-150 used the whole chassis as the heatsink.
Hi,
If you transfer more heat to the heatsink it gets hotter.
STK IC's normally don't get hot when on a heatsink except when running at full power.
I am sure the metal case did conduct much heat off the heatsink as the metal is a far more better conductor than air. Or see the metalcase as a extension of the heatsink. I am sure this is the cause for the hotter heatsink now.
Old Crown amplifier D-150 used the whole chassis as the heatsink.
Hi, Arndt
http://radvs.boom.ru/ims-unch5.html
has a little data on the STK.
http://www.datadart.com/al/sanyo/Stk075.pdf is a bad datasheet
I think STK's never run hot at idle.
Just warm should be normal.
Check their power supply.
Should be around 27-39V
Hope this can help a bit.
/Hugo - thinks more heat sink is not necessary.
http://radvs.boom.ru/ims-unch5.html
has a little data on the STK.
http://www.datadart.com/al/sanyo/Stk075.pdf is a bad datasheet
I think STK's never run hot at idle.
Just warm should be normal.
Check their power supply.
Should be around 27-39V
Hope this can help a bit.
/Hugo - thinks more heat sink is not necessary.
As has been noted, better thermal conduction from source to heatsink results in hotter heatsinks and cooler dies. The effect should not be all that pronounced, though ....
Now, if you want better cooling, the cheapest way to do it is to raise the unit above what it sits on so that you get better convection. This is very effective.
Petter
Now, if you want better cooling, the cheapest way to do it is to raise the unit above what it sits on so that you get better convection. This is very effective.
Petter
Oops, gave you the wrong link to STK080G
http://www.datadart.com/al/sanyo/Stk075g.pdf should
be better.
The recommended power supply here is 27.5V
The quiescent current is 100mA Max.
So the heat sinks can only become warm at idle.
Maybe one of your mods is the cause?
Did they get hot before the mods?
Maybe someone can point out how to measure the idle current?
/Hugo
http://www.datadart.com/al/sanyo/Stk075g.pdf should
be better.
The recommended power supply here is 27.5V
The quiescent current is 100mA Max.
So the heat sinks can only become warm at idle.
Maybe one of your mods is the cause?
Did they get hot before the mods?
Maybe someone can point out how to measure the idle current?
/Hugo
Well, the "mods" I did were as follows:
1. take out the amplfier-board, disconnecting pre-amplifier and receiver part
2. disconnect the VU-meters from the outputs
3. replace the two filtering caps TOWA 6800 uF / 42 V with BHC SlitFoil 8200 uF / 50 V
4. take out the old "wrap-the-wire-around" pins, replacing them with new soldering pins
5. replace ALL electrolytics with Panasonic FC of same value, sometimes higher voltage rating, except for 1uF / 50 V which I could not get as Pansonic FC, so I used Elna Silmic, and 47 uF / 63 V, for which I took BC components elko
6. (just incorporated) replace old heatsink grease with ArcticSilver III
Actually, all those modifications resulted in colder components except for the heasink. There is a resistor mounted on board (R806), which is packaged in a special case (so it does not desolder itself), which got very hot with the old electrolytics (I actually burned my finger once on it), which now stays relatively cool to the touch.
But I have to admit that I did not check voltages yet. I have the service manual here, but I don't know how to measure quiscent current (maybe with 1 ohm resistor in series with supply, measuring voltage across...?), and there is no trimpot at all on the board with which to set the current...
I have to say that you really got me frightened now... there are no STK080Gs available any more, and the only STK that comes close to 080G in the name I can get is a STK086, which costs 44.50 € (about 46 $)...
So, again, help to rule out any electronical faults in the circuit would be very appreciated...
Bye,
Arndt
1. take out the amplfier-board, disconnecting pre-amplifier and receiver part
2. disconnect the VU-meters from the outputs
3. replace the two filtering caps TOWA 6800 uF / 42 V with BHC SlitFoil 8200 uF / 50 V
4. take out the old "wrap-the-wire-around" pins, replacing them with new soldering pins
5. replace ALL electrolytics with Panasonic FC of same value, sometimes higher voltage rating, except for 1uF / 50 V which I could not get as Pansonic FC, so I used Elna Silmic, and 47 uF / 63 V, for which I took BC components elko
6. (just incorporated) replace old heatsink grease with ArcticSilver III
Actually, all those modifications resulted in colder components except for the heasink. There is a resistor mounted on board (R806), which is packaged in a special case (so it does not desolder itself), which got very hot with the old electrolytics (I actually burned my finger once on it), which now stays relatively cool to the touch.
But I have to admit that I did not check voltages yet. I have the service manual here, but I don't know how to measure quiscent current (maybe with 1 ohm resistor in series with supply, measuring voltage across...?), and there is no trimpot at all on the board with which to set the current...
I have to say that you really got me frightened now... there are no STK080Gs available any more, and the only STK that comes close to 080G in the name I can get is a STK086, which costs 44.50 € (about 46 $)...
So, again, help to rule out any electronical faults in the circuit would be very appreciated...
Bye,
Arndt
Well, I just measured the supply voltage. It is almost exactly 40 V. While this seems to be very high if compared to suggested voltage, it is the voltage that Marantz intended for this amp...
Btw, since the original transformer hums quite a bit, would it be OK to just use one with lower voltage, or would that force me to exchange a lot of other components as well? - And, the amp also need some additional voltages, which I would have to supply with additional toroids, here the orig transformer has additional windings...
Bye,
Arndt
Btw, since the original transformer hums quite a bit, would it be OK to just use one with lower voltage, or would that force me to exchange a lot of other components as well? - And, the amp also need some additional voltages, which I would have to supply with additional toroids, here the orig transformer has additional windings...
Bye,
Arndt
Arndt,
If I frightened you, sorry for that!
STK's usually don't have trimpots in there circuits.
Measuring the idle current could be done with a 1ohm resistor
in series with VCC at pin 8 and (or) pin 6.
If you have a scope I would check pin 1 and pin 7 for oscillations.
Careful with the probes, discharge the main caps before measuring !!!
The only strange thing I can find in your mods is the resistor.
Is it a high power resistor? Several watts?
Can you post a piece of schematic?
As to your last post: I wouldn't do any mods to the amp anymore.
First try to tame that heat.
/Hugo - won't respond anymore today, is going out to a nice little restaurant now...
If I frightened you, sorry for that!
STK's usually don't have trimpots in there circuits.
Measuring the idle current could be done with a 1ohm resistor
in series with VCC at pin 8 and (or) pin 6.
If you have a scope I would check pin 1 and pin 7 for oscillations.
Careful with the probes, discharge the main caps before measuring !!!
The only strange thing I can find in your mods is the resistor.
Is it a high power resistor? Several watts?
Can you post a piece of schematic?
As to your last post: I wouldn't do any mods to the amp anymore.
First try to tame that heat.
/Hugo - won't respond anymore today, is going out to a nice little restaurant now...
Hi!
I just scanned the schematic. Here is the picture (note that there is a strong line visible which is not a connection but a shadow from where the paper was folded):
And I only mentioned the resistor because of it producing a lot of heat before electrolytics replacement, and nearly none at all after the replacement.. The original Marantz BOM has R806 as a 150 ohms / 3 W resistor. If you have access to Farnell pictures (catalogue), it looks like FCB2 resistors from Meggitt Electronic Components, Farnell No 148-736 (for example).
Well, maybe I can hunt down some voltages using the schematic, since the "should be" values are printed at important junctions.
I don't have a scope, though, and underneath the board there are two ceramic caps directly soldered to pin 4 and 5 of the IC, maybe to prevent oscillations... they seem not to be in the schematic. Also two small electrolytics were installed underneath the board, but I don't have access to it right now, so I cannot tell exactly where...
Anyway, so far thanks for the help...
Bye,
Arndt
I just scanned the schematic. Here is the picture (note that there is a strong line visible which is not a connection but a shadow from where the paper was folded):
An externally hosted image should be here but it was not working when we last tested it.
And I only mentioned the resistor because of it producing a lot of heat before electrolytics replacement, and nearly none at all after the replacement.. The original Marantz BOM has R806 as a 150 ohms / 3 W resistor. If you have access to Farnell pictures (catalogue), it looks like FCB2 resistors from Meggitt Electronic Components, Farnell No 148-736 (for example).
Well, maybe I can hunt down some voltages using the schematic, since the "should be" values are printed at important junctions.
I don't have a scope, though, and underneath the board there are two ceramic caps directly soldered to pin 4 and 5 of the IC, maybe to prevent oscillations... they seem not to be in the schematic. Also two small electrolytics were installed underneath the board, but I don't have access to it right now, so I cannot tell exactly where...
Anyway, so far thanks for the help...
Bye,
Arndt
Neither the 32.8V nor the 14.5V seem to be used for the STK's.
I think they were used for the preamp or the receiver.
Anyone, correct me here if I'm wrong.
So I wouldn't pay much attention to R806.
You can easily measure the idle current by replacing F8(9?)02 and
F803 with 1ohm resistors.
Of course you will measure the idle current of both STK's then.
The ceramic caps will probably be there to prevent oscillation, as you mentioned.
Go ahead and measure the voltages around the IC's!
/Hugo - Think Elso could be right if all measures ok
I think they were used for the preamp or the receiver.
Anyone, correct me here if I'm wrong.
So I wouldn't pay much attention to R806.
You can easily measure the idle current by replacing F8(9?)02 and
F803 with 1ohm resistors.
Of course you will measure the idle current of both STK's then.
The ceramic caps will probably be there to prevent oscillation, as you mentioned.
Go ahead and measure the voltages around the IC's!
/Hugo - Think Elso could be right if all measures ok
Cradle22 said:
Are you sure you are measuring +/-40v at the main chip power pins? It certainly should be +/-35v or less according to the circuit.
Sanyo dataheets set +/-39v as the maximum voltage for the STK080. +/- 1v shouldn't be that serious, but you might be safer if you reduce them somehow. Go here see the data:
http://www.datadart.com/al/sanyo/Stk075.pdf
As you can see, Sanyo recommends +/-27v por 8 ohm operation, which is quite a bit lower than what Marantz specifies.
The matter deserves investigation.
Carlos
carlmart said:
I nearly misled Arndt and now it (nearly) happens again.
The STK is the STK080G so here's the correct link
http://www.datadart.com/al/sanyo/Stk075g.pdf
On that sheet you can see that the Max Voltage is 40V.
Recommended 27.5V
Voltage on the Marantz schematic: 35V
Voltage measured by Arndt: "almost exactly 40 V"
Anyway, I have to admit that there is not too much difference between the two datasheets regarding voltage
We will have to wait for other measurements…
/Hugo - Doesn’t like the letters behind transistor numbers
Good Morning!
Gosh! While looking at the schematic I accidently swallowed (well, swallowed a bit, spat out the rest)a whole mouth full of boiling hot tea prior to adding the cold milk like I'm used to...
Sometimes distractions provided by DIY electronics can be very painful...
!
I will try measurements in a little while... the "nearly 40 volts" I mentioned were measured directly after the rectifying diodes, at J818 and J819. I do not think that the voltage will change on the way to pin 6 on the STKs, where 35 V should be connected... but we will see...
Bye,
Arndt
Gosh! While looking at the schematic I accidently swallowed (well, swallowed a bit, spat out the rest)a whole mouth full of boiling hot tea prior to adding the cold milk like I'm used to...
Sometimes distractions provided by DIY electronics can be very painful...
!I will try measurements in a little while... the "nearly 40 volts" I mentioned were measured directly after the rectifying diodes, at J818 and J819. I do not think that the voltage will change on the way to pin 6 on the STKs, where 35 V should be connected... but we will see...
Bye,
Arndt
Hi!
Maybe I have found it - or maybe I just measured something wrong...
First of all, the transformer supplies a voltage of 30.2 V AC.
This comes up to about +/- 40.5 V after rectifying diodes
At pin 10 we have +39.6 V
At pin 6, where the negative voltage should be, we have 0.0 V...!!
EDIT::: second measurements showed - 40 V at pin 6... don't know why I measured nothing on first attempt... Accidentally I shorted P6 and P7 with the probes, big spark, but nothing seems to be broken... lucky...
While the fuses are working!
Don't know what could cause this...
But shouldn't the amp refuse to work at all without the negative rail?...
To quote Yoda: Confusing this gets, yes, yes...
Still looking,
Arndt
Maybe I have found it - or maybe I just measured something wrong...
First of all, the transformer supplies a voltage of 30.2 V AC.
This comes up to about +/- 40.5 V after rectifying diodes
At pin 10 we have +39.6 V
At pin 6, where the negative voltage should be, we have 0.0 V...!!
EDIT::: second measurements showed - 40 V at pin 6... don't know why I measured nothing on first attempt... Accidentally I shorted P6 and P7 with the probes, big spark, but nothing seems to be broken... lucky...
While the fuses are working!
Don't know what could cause this...
But shouldn't the amp refuse to work at all without the negative rail?...
To quote Yoda: Confusing this gets, yes, yes...
Still looking,
Arndt
Must've killed something in the additional supply...
J826 no longer carries 14.5 V, and to that pin my DC protection / speaker delay board was connected.
No fuse is blown, so I think maybe I killed some semiconductor in the addtional supply part of the schemtica. Don't know why, since I am absolutely sure that I "only" shorted PIN 6 and 7 of the IC, and only P1 and P2 are somewhat connected to the additional supply circuit...
Re-soldering the speaker cables right now, and sending out prayers to heaven that everything will work out alright...
Getting back when done,
Arndt
J826 no longer carries 14.5 V, and to that pin my DC protection / speaker delay board was connected.
No fuse is blown, so I think maybe I killed some semiconductor in the addtional supply part of the schemtica. Don't know why, since I am absolutely sure that I "only" shorted PIN 6 and 7 of the IC, and only P1 and P2 are somewhat connected to the additional supply circuit...
Re-soldering the speaker cables right now, and sending out prayers to heaven that everything will work out alright...
Getting back when done,
Arndt
OK, I'm done - with everything!
Spend quite some money on those Panasonic FCs and BHC SlitFoil, and now I've killed the left amp... Buah, and this was the best sounding amp I ever had... better than my GainClone...
There is a loud hum coming out of the left speaker when turning on the amp.
But maybe (hope, hope) the fault isn't in the main IC, or do you think I'm not that lucky...???
If I've killed the STK that would be the fourth IC I've killed by slipping with the probes (three LM3876 were the first victims)...
I do not have very steady hands...
Bye,
Arndt
Spend quite some money on those Panasonic FCs and BHC SlitFoil, and now I've killed the left amp... Buah, and this was the best sounding amp I ever had... better than my GainClone...
There is a loud hum coming out of the left speaker when turning on the amp.
But maybe (hope, hope) the fault isn't in the main IC, or do you think I'm not that lucky...???
If I've killed the STK that would be the fourth IC I've killed by slipping with the probes (three LM3876 were the first victims)...
I do not have very steady hands...
Bye,
Arndt
Cradle22 said:
Jesus...
Disconnect the speakers from the amp (He might carry DC!) and look if both +&- 35V are there.
If they are, and probably they will be there because the right channel works, the STK might be gone.
It's not your lucky day; first the tea, then this.
/ Hugo - Thinks tele-repair is difficult.
Cradle22 said:
One of the best test gear investments I ever made was to equip my meter with a set of Pomona Model 6469 SMD Microtip Test Probes. These have spring-loaded pogo-pin tips that don't slip off the component leads when you look away from the circuit. The tips are like tiny tridents and bite slightly into the metal. Way cool, you can wiggle the probe around and the tip doesn't slip. A set of leads cost less than $20 and the tips are replacable.
Pomona also makes scope probes with the same spring loaded tips that are great for making measurements on tight pitched components. I am using some 0.5mm pitch components on my new designs and the probes that came with my Tek scope aren't worth beans for that.
http://www.pomonaelectronics.com/pdf/d6469_1_02.pdf
Phil
Hi!
Well, the voltage was there, so the IC must be broken. I disconnected the broken channel from the supply so that I can still use the other one...
In Germany there is a Marantz SR2000 (the one I ripped apart) on Ebay, but reported as broken with DC on the speaker connectors. That very much indicates broken ICs, as well, doesn't it? So bying that one wouldn't help me, or could that also be another fault?
Re-assembling my GainClone this evening so that I will have a working amp...
Bye,
Arndt
Well, the voltage was there, so the IC must be broken. I disconnected the broken channel from the supply so that I can still use the other one...
In Germany there is a Marantz SR2000 (the one I ripped apart) on Ebay, but reported as broken with DC on the speaker connectors. That very much indicates broken ICs, as well, doesn't it? So bying that one wouldn't help me, or could that also be another fault?
Re-assembling my GainClone this evening so that I will have a working amp...
Bye,
Arndt
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.