Progressing along nicely, all resistors are stuffed and soldered, next order of parts should be here tomorrow.
I was getting ready to polish up the heatsink and strip the anodizing so as to improve the thermal transfer, then I thought why not just use grease and direct mount to the heatsink, I think the anodizing will have enough isolation for 30VDC?
Could someone more experienced in this than I confirm I can mount directly on the heatsink with just grease?
Cheers
I was getting ready to polish up the heatsink and strip the anodizing so as to improve the thermal transfer, then I thought why not just use grease and direct mount to the heatsink, I think the anodizing will have enough isolation for 30VDC?
Could someone more experienced in this than I confirm I can mount directly on the heatsink with just grease?
Cheers
You can, but it's somewhat risky.
The aluminium under the anodising might also be ridged, so I prefer to flatten it (not necessarily all the way through the anodising, but enough so that it's flat) and then use grease and mica. YMMV.
The aluminium under the anodising might also be ridged, so I prefer to flatten it (not necessarily all the way through the anodising, but enough so that it's flat) and then use grease and mica. YMMV.
😱 Don't risk it, it's guaranteed to fail sooner or later, Sage Audio made that mistake!
Pass DIY Addict
Joined 2000
Paid Member
+1 Do things the "right way" (with an electrical insulator) and you won't regret it later 😉 If you use some very fine sand paper and a flat piece of metal, you can polish the heatsink mounting site and you'll get better thermal transfer. Lately, I've been using ceramic insulators with heatsink grease. They perform better than mica and goop and are much more robust (physically) than any SilPad.
Last edited:
Okay thanks guys, much appreciated. I will do the right thing. I have Berquist SilPad K10, one of their top quality Kapton on the way. I will polish the areas where the FETs are going, I think now it is best to remove the anodizing so it's down to clean aluminum, then the pads will have a nice flat surface to go on.
Starting to work on the layout and power supply now as I am waiting for more parts to arrive. I am torn between using 2 torroids, 20VAC dual windings 5A each, or one 10A dual 20V windings. I have both and have lots of large 27,000uF caps with screw terminals, will need to do a lot of drilling to mount all the clamps. Other option is to order the PS board from the DIY store but that will add 3 weeks onto the build, and all new caps. Mmmmm? What to do. 😕
Is it detrimental to use just one large supply rather than true dual mono in an amp of this size?
Thanks
PS I would have used mica and grease but I can't find my mica insulators in the 20 boxes of stuff I have, and Digikey doesn't seem to carry them anymore.
Starting to work on the layout and power supply now as I am waiting for more parts to arrive. I am torn between using 2 torroids, 20VAC dual windings 5A each, or one 10A dual 20V windings. I have both and have lots of large 27,000uF caps with screw terminals, will need to do a lot of drilling to mount all the clamps. Other option is to order the PS board from the DIY store but that will add 3 weeks onto the build, and all new caps. Mmmmm? What to do. 😕
Is it detrimental to use just one large supply rather than true dual mono in an amp of this size?
Thanks
PS I would have used mica and grease but I can't find my mica insulators in the 20 boxes of stuff I have, and Digikey doesn't seem to carry them anymore.
Last edited:
One PSU works beautifully. 
Dial mono in a single case isn’t worth the bother in my opinion, make monoblocks for actual benefit.
Dial mono in a single case isn’t worth the bother in my opinion, make monoblocks for actual benefit.
Hi Jim,
Thanks for that, good to see you in this thread again and thank you for all of your work with this project, and other projects as well.
🙂
Well that's a good thing as I just started to layout the bottom and there is no way to get all the caps for dual xfmrs in there, it will be a challenge to get 8 in there. 🙄 Now that I am thinking about it, the 5A xfmr actually would be enough for both channels as each channel only draws ~1.7A DC. I also have a plethora of dual 24VAC @ 8A dual winding torroids as well, but that would cause excessive heat in the amp. I think I might save the big 10A torroids for my F5T build.
Anyway, thank you again.
Cheers
Thanks for that, good to see you in this thread again and thank you for all of your work with this project, and other projects as well.
🙂Well that's a good thing as I just started to layout the bottom and there is no way to get all the caps for dual xfmrs in there, it will be a challenge to get 8 in there. 🙄 Now that I am thinking about it, the 5A xfmr actually would be enough for both channels as each channel only draws ~1.7A DC. I also have a plethora of dual 24VAC @ 8A dual winding torroids as well, but that would cause excessive heat in the amp. I think I might save the big 10A torroids for my F5T build.
Anyway, thank you again.
Cheers
Greetings, all!
I'm finally to the point of initial power up on my Aleph J. I'm testing the PSU currently, and one of my LEDs is not lighting up. However, I'm getting a steady 25v on the rails. Do I need to worry about trouble shooting the LED at this point or can I proceed to hooking up the Aleph boards?
Note: I'm using the lightbulb tester with a 100w bulb. It lit up brightly for a second and dimmed as expected. I noticed the LED flash a couple times when I first supplied power, but then it went out. The other PSU LED lit up and is holding steady.
I'm finally to the point of initial power up on my Aleph J. I'm testing the PSU currently, and one of my LEDs is not lighting up. However, I'm getting a steady 25v on the rails. Do I need to worry about trouble shooting the LED at this point or can I proceed to hooking up the Aleph boards?
Note: I'm using the lightbulb tester with a 100w bulb. It lit up brightly for a second and dimmed as expected. I noticed the LED flash a couple times when I first supplied power, but then it went out. The other PSU LED lit up and is holding steady.
Last edited:
DVM is smarter than LED , so thrust in DVM
you can rotate LED later or right away , your choice
you can rotate LED later or right away , your choice
^^ Yes, what he said.
Your LED is most likely in backwards, flip it around if you want. Or not. 🙂
Your LED is most likely in backwards, flip it around if you want. Or not. 🙂
Okay, so on the right channel I've got the bulb connected and am working with the DC offset. I've got it down to 0.118V. Bulb did not dim at all this time, but based on posts 2062-2064 in this thread it sounds like that's okay?
Proceed to full power and biasing?
Proceed to full power and biasing?
bulb is OK just for first testing - to prevent BigBadaBoom
then you must remove bulb for setting procedure, or it will act as nasty limiter as you are trying to bias it more
then you must remove bulb for setting procedure, or it will act as nasty limiter as you are trying to bias it more
How many watts is your light bulb? I keep a 150W in my base for most amps, I have a 100W for small amps or preamps, and a European 240V 150W for the BIG AMPs. As you are biasing a ~80W amp channel, if you are only using a 100W bulb, you will have ~1/2 the voltage going to the power supply. So as ZM said, you had no Kaboom, so you can either go to a bigger bulb or straight AC. Keep us posted.
Cheers
Pass DIY Addict
Joined 2000
Paid Member
Even with the light bulb limiter in place, you should be able to connect an input source and hear *something* on an old test speaker hooked to the speaker output. My Aleph-J made some noise/hum/hiss/distortion when used with a dim bulb, but I could clearly hear music as well. If you are hearing music, it is working and you should be able to go to full power without incident...
problem with bulb tester in circuit while setting is not so critical ........ worst thing is that one probably is not going to set it
problem can arise later - amp tried to be set with bulb tester in , can go berserk when you remove bulb tester , forgetting to back down everything
example - try setting F5 with bulb tester in ....... you can set DC offset soso , but Iq not , no matter how much you dial trimpots
remove bulb tester - BigBadaBoom
problem can arise later - amp tried to be set with bulb tester in , can go berserk when you remove bulb tester , forgetting to back down everything
example - try setting F5 with bulb tester in ....... you can set DC offset soso , but Iq not , no matter how much you dial trimpots
remove bulb tester - BigBadaBoom
Pass DIY Addict
Joined 2000
Paid Member
From time to time, I entertain the idea of building a new little contraption for those "first power up" moments. I'm thinking of putting my variac at the front end, to be followed by a selectable set of other items:
1) an analog AC voltmeter - the meter on my variac is pretty poor and the scale printed under the dial is rather worthless in terms of precision. I can determine "more voltage" and "less voltage" and sometimes even "ballpark voltage" but the scale is pretty inaccurate.
2) a dim-bulb tester to look for egregious problems, and
3) an analog current meter (5-10A range) so I can actually see what kind of current draw is happening, how it behaves upon powerup, and how it changes when the PSU reaches steady state. This is harder to see with my DMM as the digital display jumps all over the place in the first second or two before things settle down.
I keep thinking this will provide more insight as to what is happening as I power up my next project...
1) an analog AC voltmeter - the meter on my variac is pretty poor and the scale printed under the dial is rather worthless in terms of precision. I can determine "more voltage" and "less voltage" and sometimes even "ballpark voltage" but the scale is pretty inaccurate.
2) a dim-bulb tester to look for egregious problems, and
3) an analog current meter (5-10A range) so I can actually see what kind of current draw is happening, how it behaves upon powerup, and how it changes when the PSU reaches steady state. This is harder to see with my DMM as the digital display jumps all over the place in the first second or two before things settle down.
I keep thinking this will provide more insight as to what is happening as I power up my next project...
Variac with built in voltmeter, plus AC ammeter, plus very cautious electronic technician, are sufficient.