K-amps said:I just mentioned it after I saw you dis-assemble the entire 565.... I said to myself, there's someone more crazy than I am...
LOL! Hahaha thanks for the laugh!
Yes I tend to be very obsessive about things as I like to focus 100% into single goals, and right now my goal is to make the inside of the GFA-565 a work of art. I want all of the wires to run on little mini-highways inside of a chassis where I will custom cut the RFI/EMI shielding foil to affix onto.
It's the equivalent of detailing a car, but replacing all of the critical components too.
When I first opened the 565's top cover and saw THIS I literally sat down and held my head, what a disgusting wiring mess and poor soldering skills too on the non-machine solder points. Once I have the unit reassembled every 565 out there will be wanting some MJR AUDIO love!
BTW, THIS gentleman obviously shares my design standards.
Thank you Arif and Chris so much for all your help I am very fortunate to have met you two on diyAudio.com. This site is a 10/10 even though I'm sitting on
anatech said:
This morning I discovered how the fan circuit operates.
The fan power supply PCB takes a 21VAC input from one of the 565 toroid's secondaries, then rectifies that voltage and regulates it into a DC signal which constantly varies in voltage based on the temperature that the thermistor is feeding it.
Using a blow-dryer aimed directly at the thermistor (which was outside of the amp during this test), I found that the fan engaged at ~65 degrees Celsius and shut off at around 50 degrees Celsius. As the temperature of the thermistor rose, the VDC output to the fan increased and the fan spun faster, delivering more air.
Based on these findings and my experience with using the amps with a 8ohms load, I can conclude that in a normal operating environment, the fans will ONLY engage when the output transistor heatsink reaches 65 degrees Celsius, which can usually only come when driving a very low-impedance load at intense output levels.
I can either replace the thermistor with a unit that actuates at a lower temperature, or I can simply place it on the side of one of the predriver transistors, which will always be operating at 70 degrees Celsius. The high temperature of the predrivers will initiate the fan to run continuously, cooling the entire amplifier and contributing to a more stable and ultimately longer-lasting unit.
The drawbacks we have forseen may likely be an increase in LF noise, and the possibility of predriver transistors sounding "bad" due to a lowered operating temperature, however, the latter assumption is open to debate. I argue that the sound of the actual fan running will negate any difference in output sound, which may be the most significant drawback to this modification: having to hear the fan operating!
I would not mess with the thermistor settings...
I have tried it once on another amp, it requires a lot of trial and error.... I ended up blowing up the amp twice when the Thermistor became loose or did not contact well...
It's probably not worth doing. The Trannies are supposed to get warm, in the other Adcoms, Nelson had thermal breakers set at 85c. I have ran one hard enough for it to thermal break... the amp ran fine after re-setting... if 85c is fine , 65 is a no brainer.
I have tried it once on another amp, it requires a lot of trial and error.... I ended up blowing up the amp twice when the Thermistor became loose or did not contact well...
It's probably not worth doing. The Trannies are supposed to get warm, in the other Adcoms, Nelson had thermal breakers set at 85c. I have ran one hard enough for it to thermal break... the amp ran fine after re-setting... if 85c is fine , 65 is a no brainer.
K-amps said:
The fan circuit comes with its own dedicated thermistor, used only as part of the fan circuit while the amp's thermo-controllers are UP62 Uchiya Thermal Breakers which close at 86 degrees Celsius, trigguring the thermal protection circuitry on the input PCB, I'm not moving those thermal cutoffs but merely moving the fan circuit's thermistor, which is independent of the amp's thermal protection circuitry.
How did you blow up the amp? Overheating caused by overdriving?
I blew the amp cause I did'nt know better...
I had thermistors in close contacts with the sinks bit not affixed, as the fan blew, it cooled the thermistor though did not the sink as much and the sink got hotter and hotter. The fan would recycle on/off as it cooled the thermistors... it was funny... I was like, Mr. Fan your foreplay needs to involve the heatsink, don't stop after the thermistor has cooled off...
With no Bias diodes on the heatsink, they went into thermal runaway even without a signal. It was a bad amp design, but the point is, know the consequences of your actions...
I had thermistors in close contacts with the sinks bit not affixed, as the fan blew, it cooled the thermistor though did not the sink as much and the sink got hotter and hotter. The fan would recycle on/off as it cooled the thermistors... it was funny... I was like, Mr. Fan your foreplay needs to involve the heatsink, don't stop after the thermistor has cooled off...
With no Bias diodes on the heatsink, they went into thermal runaway even without a signal. It was a bad amp design, but the point is, know the consequences of your actions...
Rebuild Pictorial
Gentlemen,
I have been busy working on the 565, completely rewiring it and adding modularity to the unit's circuitry while trying to retain and/or surpass factory specifications. I have made many modifications to the unit and I believe it should perform superbly now. I am waiting on more Molex Gold female connectors and new 1.8V 7mA low-current green LEDs for the input stage, since the 1.8V needle-style originals are en route to failure.
I am hoping to have the amp done by Saturday.
Enjoy the pictorial!
PIC 1: Input Stage after 1 hour of solid ultrasonic cleaning, completely rebuilt with brand new Mallory SEK series 105-degree-rated aluminum electrolytic capacitors, WECO PCB screw-down terminals (epoxied to the PCB with cyanoacrylate, then soldered), gold headers for all LED connection points, an upgraded Analog Devices OP97 op-amp IC, a Copal 100 ohms Cermet variable resistor, and silver-plated copper patch wires with teflon dielectric.
PIC 2: Photo of the other side of the board showing the new OP97 op-amp, Mallory electrolytics, and regreased Sanyo 150W Predriver Transistors (using Arctic Silver 5 thermal compound).
PIC 3: The GFA-565 input stage PCB was bathed through 3 Simple Green washes and 4 cold water rinses inside of this ultrasonic cleaner, yeilding a wash cycle of 1 full hour. The tank was drained, rinsed, and wiped-down after each wash cycle, to prevent any remaining electrolyte from reattaching to the PCB.
PIC 4: Modified Optional Top Cover & Cooling Fan (EMI/RFI Aluminum/Polyester Shielding Foil installed for over 60dBA attenuation of 600MHz-1GHz frequencies, upgraded bypass capacitors to 105 degree Celsius-rated aluminum electrolytics & custom gold-plated copper headers installed with brass-female connectors crimped into the AC input and DC output leads for modular connections. In addition, the thermistor for the fan circuit is placed on the input stage predriver transistors, which force the fan to flush out the hot air every 5 minutes, environmental noise from the fan is negligible at normal listening levels.)
Gentlemen,
I have been busy working on the 565, completely rewiring it and adding modularity to the unit's circuitry while trying to retain and/or surpass factory specifications. I have made many modifications to the unit and I believe it should perform superbly now. I am waiting on more Molex Gold female connectors and new 1.8V 7mA low-current green LEDs for the input stage, since the 1.8V needle-style originals are en route to failure.
I am hoping to have the amp done by Saturday.
Enjoy the pictorial!
PIC 1: Input Stage after 1 hour of solid ultrasonic cleaning, completely rebuilt with brand new Mallory SEK series 105-degree-rated aluminum electrolytic capacitors, WECO PCB screw-down terminals (epoxied to the PCB with cyanoacrylate, then soldered), gold headers for all LED connection points, an upgraded Analog Devices OP97 op-amp IC, a Copal 100 ohms Cermet variable resistor, and silver-plated copper patch wires with teflon dielectric.
PIC 2: Photo of the other side of the board showing the new OP97 op-amp, Mallory electrolytics, and regreased Sanyo 150W Predriver Transistors (using Arctic Silver 5 thermal compound).
PIC 3: The GFA-565 input stage PCB was bathed through 3 Simple Green washes and 4 cold water rinses inside of this ultrasonic cleaner, yeilding a wash cycle of 1 full hour. The tank was drained, rinsed, and wiped-down after each wash cycle, to prevent any remaining electrolyte from reattaching to the PCB.
PIC 4: Modified Optional Top Cover & Cooling Fan (EMI/RFI Aluminum/Polyester Shielding Foil installed for over 60dBA attenuation of 600MHz-1GHz frequencies, upgraded bypass capacitors to 105 degree Celsius-rated aluminum electrolytics & custom gold-plated copper headers installed with brass-female connectors crimped into the AC input and DC output leads for modular connections. In addition, the thermistor for the fan circuit is placed on the input stage predriver transistors, which force the fan to flush out the hot air every 5 minutes, environmental noise from the fan is negligible at normal listening levels.)
Hi mjraudio,
BTW, errors in that article. Anyone can put anything up on the web.
Okay, I guess if it was hanging around or easier for you to get. Everyone should stick to normal heatsink compound. Is this stuff conductive? I don't know.
You are thermal cycling everything! Why not run the fan continuously at lower RPM? I think you should have stopped with the new heatsinks on the drivers. You also really don't want air flow. I guess the collected dust in your new vacuum cleaner will solve that one for you eventually.
-Chris
No they aren't. Just make sure the voltage drop is the same as the originals were. They aren't there to emit light either.
BTW, errors in that article. Anyone can put anything up on the web.Unwise. The plastic packages are not rated for that kind of abuse. Issues may not show up for years though. I wish you luck, don't repeat this on anything else for fear of inducing failure mechanisms.
Okay, I guess if it was hanging around or easier for you to get. Everyone should stick to normal heatsink compound. Is this stuff conductive? I don't know. You are thermal cycling everything! Why not run the fan continuously at lower RPM? I think you should have stopped with the new heatsinks on the drivers. You also really don't want air flow. I guess the collected dust in your new vacuum cleaner will solve that one for you eventually.That's good.
I hear they make the military foil hats from this stuff to keep the CIA from implanting thoughts. Why do you need it? It's not much good without the gaskets, feed through caps and line filters.
These are not always better. I've measured many 85° caps outperforming the 105 ° types. Their doesn't seem to be any rhyme or reason to this. Bypass them with smaller foil types.
-Chris
Hmm two of your comments have got me worried:
First, the overcleaning I did on the board....what packages are not rated for that kind of heavy cleaning? The diodes and mylar capacitors? I hope I did not injure anything!
And second: the choice of 105-degree Mallory electrolytics...I understood that all of the electrolytics in the 565 were using for DC coupling and/or as a bypass caps in the circuits...How do electrolytics "outperform" others in their application within the 565 and also how could it benefit from higher quality capacitors such as 85-degree rated or even Black Gate capacitors? I chose 105-degree for the safety of having a higher-temp rated cap....I still am not comfortable with 70-degree predrivers on a little board with electrolytic capacitors right next to their heatsinks. Hmm maybe I jumped the gun of cap. choice here...
Also, I ended up not changing the heatsinks on the predrivers, upgraded TO-220 heatsinks would simply not fit on the small input stage PCB, the originals went back on.
Your advice is much appreciated as always Chris!
First, the overcleaning I did on the board....what packages are not rated for that kind of heavy cleaning? The diodes and mylar capacitors? I hope I did not injure anything!
And second: the choice of 105-degree Mallory electrolytics...I understood that all of the electrolytics in the 565 were using for DC coupling and/or as a bypass caps in the circuits...How do electrolytics "outperform" others in their application within the 565 and also how could it benefit from higher quality capacitors such as 85-degree rated or even Black Gate capacitors? I chose 105-degree for the safety of having a higher-temp rated cap....I still am not comfortable with 70-degree predrivers on a little board with electrolytic capacitors right next to their heatsinks. Hmm maybe I jumped the gun of cap. choice here...
Also, I ended up not changing the heatsinks on the predrivers, upgraded TO-220 heatsinks would simply not fit on the small input stage PCB, the originals went back on.
Your advice is much appreciated as always Chris!
A few 85ºC caps perform better than many 105ºC caps, but pretty much any new cap will out-perform the originals. Don't sweat this one too much...what you have is just fine. One advantage of the 105°C caps is much longer lifespan. Personally, I use 105°C Panasonic FM and FC caps. Tough to beat, and they'll likely outlive me.
Most modern parts are rated for immersion in liquids for cleaning, but long periods in a surfactant, with ultrasonics no less...hmmm. My thinking here is that if it powers up and works, you're fine. You'll know soon enough.
Most modern parts are rated for immersion in liquids for cleaning, but long periods in a surfactant, with ultrasonics no less...hmmm. My thinking here is that if it powers up and works, you're fine. You'll know soon enough.
Hi mjraudio,
Glenn is right when he says any new electrolytic capacitor will out perform most old ones. I find sometimes that the better capacitors in old Marantz (and others I'm sure) gear may out perform the new parts - after 30 years! I test every part before replacing it on an HP 4263A LCR meter. Dissipation is where you can really see the difference. I will not replace a good cap that measures better than the new ones.
Go for a continuous but reduced airflow from the fan if you must, but return the thermal sensor to it's original location. Do not cycle the fan on and off. Filter the air if you can.
-Chris
All of them. The transistors use an epoxy to metal seal that can be compromised. Fluid + vibration isn't good. Capacitors are even more susceptible to seal failure. None of these issues will show up right away.
Glenn is right when he says any new electrolytic capacitor will out perform most old ones. I find sometimes that the better capacitors in old Marantz (and others I'm sure) gear may out perform the new parts - after 30 years! I test every part before replacing it on an HP 4263A LCR meter. Dissipation is where you can really see the difference. I will not replace a good cap that measures better than the new ones.
That's really too bad. One upgrade that I really liked.
Go for a continuous but reduced airflow from the fan if you must, but return the thermal sensor to it's original location. Do not cycle the fan on and off. Filter the air if you can.
-Chris
Thanks for your posts.
@Glen: What aspects of a capacitor's performance let an 85 degree rated cap "perform better" than its 105-degree rated counterpart? I am wondering if it could be tolerance ratings, ESR value, or capacitance value...any thoughts?
@Arif: Yes they certainly are! I just hope they are not damaged, we will see.
@Chris: Constant thermistor actuations will probably wear out the fan psu in addition to the thermistor itself, I think you are right. I will try to fix the fan at a continuous speed, thanks.
@Glen: What aspects of a capacitor's performance let an 85 degree rated cap "perform better" than its 105-degree rated counterpart? I am wondering if it could be tolerance ratings, ESR value, or capacitance value...any thoughts?
@Arif: Yes they certainly are! I just hope they are not damaged, we will see.
@Chris: Constant thermistor actuations will probably wear out the fan psu in addition to the thermistor itself, I think you are right. I will try to fix the fan at a continuous speed, thanks.
Hi mjraudio,
-Chris
Dissipation, which is related to high frequency performance. You want them to work well (as well as they can) at all frequencies. Bypassing will help with this.
It's the thermal cycling that make make your offsets drift around so that the amp is never really settled. You would have to monitor the servo output to see this (don't). I'm not sure if there would be a sonic penalty.
-Chris
Arctic Silver 5
from mfg's website:
"Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)"
underlining added by me...
from mfg's website:
"Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)"
underlining added by me...
I took the necessary precautions to make sure no trace of it touches the leads of the predrivers, but I'll clean them again and consider changing it.
I had a two tubes of that stuff sitting in my drawer for the past few years, thought I'd put it to use.
I've asked this before, but I am still wondering if I could benefit from Black Gate electrolytic capacitors on the input stage...any thoughts?
I had a two tubes of that stuff sitting in my drawer for the past few years, thought I'd put it to use.
I've asked this before, but I am still wondering if I could benefit from Black Gate electrolytic capacitors on the input stage...any thoughts?
Hi mjraudio,
Since you've got it, it should be okay in your application. Just leave it but make sure it's cleaned up around the edges.
Consider bypassing with a good film cap.
-Chris
Since you've got it, it should be okay in your application. Just leave it but make sure it's cleaned up around the edges.
In my view, nope. Complete waste of money, but they are at the top of the "modders secret hit list of must haves".
Consider bypassing with a good film cap.-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.