Thanks, Zman and Hugh. It’s a bit tight in there. That is what a pathfinder activity like this tells me. Couple of options is to move the PSU caps and bridges to the amp
PCB on the heatsink - a la Aspen.
That would free up a lot of space. We did this before on Deltic.
A custom 400VA trafo with 4x 21v secondaries would also reduce mounting points. Right now +/-30.4v rails will probably make close to 49w into 8ohms. The chassis os running a bit hot for my tastes as this is 108w dissipation per side! With the lid on it would be too much. Although a small Noctua fan would keep insides pretty cool if vented up and out.m
I love how the heatsinks can be removed or installed while amp is fully connected and running with 4 socket cap screws from the outside! Actually, all panels can be removed from the outside and this will make installing wires for the front panel much easier.
PCB on the heatsink - a la Aspen.
That would free up a lot of space. We did this before on Deltic.
A custom 400VA trafo with 4x 21v secondaries would also reduce mounting points. Right now +/-30.4v rails will probably make close to 49w into 8ohms. The chassis os running a bit hot for my tastes as this is 108w dissipation per side! With the lid on it would be too much. Although a small Noctua fan would keep insides pretty cool if vented up and out.m
I love how the heatsinks can be removed or installed while amp is fully connected and running with 4 socket cap screws from the outside! Actually, all panels can be removed from the outside and this will make installing wires for the front panel much easier.
Last edited:
I don’t know how you guys build chassis with neat wiring with 90deg turns all wrapped up? Maybe more space inside would help? It’s tight on this one and I work quickly to see if it works, thermal dissipation is ok, and parts not too close where there is noise pickup. All that is good except a bit too much heat. I grabbed these trafos from another project. Without the SLB to drop 3v, the trafo voltage will need to go down. I think 21v will be about right. With 23v, I have 35v no load and 30.4v under load.
I moved it over to my main system and on the big rack now. Connected to XSD dipoles. I had the same amp here before but it was not dual mono. I think there is a substantial improvement in soundstage width and image depth now. The channel separation makes all the difference.
Heatsinks are running at 57C.
Heatsinks are running at 57C.
I can’t believe how hot 57C is. It’s very uncomfortable to touch for more than 1 second. It is amazing that that much heat is coming from two little TO-247’s - glad I am using ceramic insulator pads with thermal grease. They can take some high heat flux vs silicone. These trafos definitely need their voltage dropped by maybe 2v down from 23v. But, I am enjoying the extra headroom in peak power.
I am right at the limit near 140F. This is a pathfinder amp with a hot 23v trafo. I’ll drop to 21v for production.
So it's the heatsink...
Wonder how hot the transistor (and the junction) is .
Must be slowly melting..
I have a ceramic insulator pad and good thermal grease. I’m sure it’s fine as 57C is only 2 above max 55C recommended for heatsinks. I like to operate at 45C if possible and it will get there once I drop the voltage to 27v. I’ll try to measure the transistor case (assuming the PCB snubber board is similar in temp as it is the clamp).
X,
A good option might to be a larger mosfet package, a hockey puck. This will offer about three times the active area and for the same dissipation it might bring the temperature down five or six degrees. This could make all the difference.
OTOH I checked availability and most of the ISOTOPS are starting at $USD30, so it could be better to simply double up on the outputs using the TO3P or TO247 case. Two mosfets in parallel should not be badThis might bring down temperatures. Sad to have to reduce the voltage just because of the dissipation issues when it sounds so good.........
HD
A good option might to be a larger mosfet package, a hockey puck. This will offer about three times the active area and for the same dissipation it might bring the temperature down five or six degrees. This could make all the difference.
OTOH I checked availability and most of the ISOTOPS are starting at $USD30, so it could be better to simply double up on the outputs using the TO3P or TO247 case. Two mosfets in parallel should not be badThis might bring down temperatures. Sad to have to reduce the voltage just because of the dissipation issues when it sounds so good.........
HD
Actually the voltage reduction is to reduce overall dissipation that the physical heatsink can handle. It’s only a 300mm deep x 150mm H unit per side. I suppose bias current could be reduced a bit? It’s at 1.8A right now. Would 1.5A work?
I just measured the temperature of the MOSFET snubber board in contact with the MOSFET body and serves as a clamp. It was 68C to 71C depending on where you aim the IR thermometer. I am not too worried about 71C for the MOSFET body temp.
I just measured the temperature of the MOSFET snubber board in contact with the MOSFET body and serves as a clamp. It was 68C to 71C depending on where you aim the IR thermometer. I am not too worried about 71C for the MOSFET body temp.
Last edited:
I’m really enjoying the Alpha Nirvana in dual mono. Track by Liz Wright sounds fantastic. The smoothness of Class A is always a delight to find again after listening to other amps.
I am new to DIY and was planning to build Pass ACA amp and procured two Meanwell 24v 5A 120W SMPS for mono blocks setup.
My Speakers are Dali Blue 6006 Floor standers of 4 Ohm and Sensitivity 91 db.
My friend suggested to go for bit more powerful class A amp and I came across this thread.
Now, I want to build AN amp, but I have some doubts as follows:-
Thanks.
My Speakers are Dali Blue 6006 Floor standers of 4 Ohm and Sensitivity 91 db.
My friend suggested to go for bit more powerful class A amp and I came across this thread.
Now, I want to build AN amp, but I have some doubts as follows:-
- Can I use Meanwell 24v 5A 120W SMPS for building AN amp to drive my existing speakers? (I cant return SMPS due to location related issues)
- Shall I go for the 4R BOM attached to post 1 for building 4 ohm version?
- PCB is same for both 4R and 8R version?
Thanks.
vbutani,
Some quick answers to your questions:
1. The AN amplifier will need an power supply with +/- voltages; hence the 24V 5A power supplies will not work.
2. The Dali 6006 is nominally rated @ 4 Ohms; I have not been able to locate an impedance curve yet for them, and that would give a better idea if there is anything else to consider. But as a minimum, you should opt for the 4 Ohm build. I cannot comment on the bill of materials for 4 Ohm version in post # 1, maybe X, Hugh, or one of the other members can chime in.
3. The PCB is the same for both the 4R and the 8R version. On post # 1 it is also mentioned that the voltage rails are lower for the 4R version (+/- 22V for the 4R vs +/- 27V for the 8R).
If I may ask, where are you located at?
Some quick answers to your questions:
1. The AN amplifier will need an power supply with +/- voltages; hence the 24V 5A power supplies will not work.
2. The Dali 6006 is nominally rated @ 4 Ohms; I have not been able to locate an impedance curve yet for them, and that would give a better idea if there is anything else to consider. But as a minimum, you should opt for the 4 Ohm build. I cannot comment on the bill of materials for 4 Ohm version in post # 1, maybe X, Hugh, or one of the other members can chime in.
3. The PCB is the same for both the 4R and the 8R version. On post # 1 it is also mentioned that the voltage rails are lower for the 4R version (+/- 22V for the 4R vs +/- 27V for the 8R).
If I may ask, where are you located at?