• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Alpha Nirvana 39W SE Class A Amplifier GB

I actually have two snubbers in place. 7mV is very good.


Ya, I wasn't worried about the DC, it is so low, I was just curious that the variable didn't seem to do anything. I measured the resistance across it and worked it from about 15R to 50R and it didn't move on the output. It should be dead quite it those levels.



Just FYI, there were only very slight changes in the outputs with the AN board snubbers jumped. Also, it changed the DC offset not at all.


Also, I shout out to Hugh and the design team, this sucker is stable as hell. I haven't made any sound with it yet, there's no hurry as I'm, very much, enjoying the J2 clone. JP wonderful work for as usual, after looking at other layouts that looked like they followed a drunkard, it's really a pleasure to build your layouts. It's especially helpful to us FNGs trying to learn.


This is going to be a mono design and the guy I got my case from seems to be taking an extra long vacation, so I don't know when I can build the other channel. I tested both in this box, so it will be ready to go whenever I can get a case. I can't find anyone who has another like this. He was supposed to be back from vacation on the 31st... Now we are 8 days past that. It's like a Mark L clone, but with the different front. If anyone has one and wants to part with it I'd be interested.




Funny story, I just got the bench supply in... I was going to power the AN with it to make it easier to work on. That would have told me it was a supply issue when it didn't current limit on the bench supply. That's moot now. :rolleyes: I'm still very glad to have one in the cave for future projects.


I still need to get back in and snub the transformer, seems Andy was right about my build pace. :nod:
 
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
If you are using SLB, snubber is not needed. There won’t be any benefit to having one really because the active bridge in the SLB switches on and off at the zero crossing. Diode bridges switch at 0.6v, and thus generates ringing pulses due to switching while current is flowing. We have snubbers installed at the AC inputs of the SLB, but that was a “belt and suspenders” thing.
 
If you are using SLB, snubber is not needed. There won’t be any benefit to having one really because the active bridge in the SLB switches on and off at the zero crossing. Diode bridges switch at 0.6v, and thus generates ringing pulses due to switching while current is flowing. We have snubbers installed at the AC inputs of the SLB, but that was a “belt and suspenders” thing.


That works for me X, I'm about done for the day, in fact I'm cooked, been at it about 9hrs and my old eyes are protesting.



Thanks again everyone, when I get some sound out of it, I will report back for you X.


Oh, I have a soft start in this one, and will probably add the speaker protection as well.



JT
 
I decided to fire her up today and put one speaker on... Absolutely quiet on 98db Tekton. I stuck my ear on the speaker and hear nothing. Keep in mind these speakers are closer to 4ohm that 8ohm. These monos will reside in my listening room on 8ohm speakers, but I wanted to do a test here. The sound great until I push the volume a bit and the impedance mismatch really starts to show.


What is the max safe rail on these X. The schematic call for 27, but what do we have either way of that, for normal operation.


For now, I will reserve comments on the sound until 1, I get it on 8ohm speakers, and 2, get the other mono built.



I'm really torqued that I can even order the second chassis yet. Prob be 2 weeks or more before I can get one here.



I guess I'll just order one from the DIY store put it in there and test it, then move it to the other chassis when it gets here. I have a couple more builds sitting here anyway.
 
Thanks Zman


X, I don't really need to crank it up, I was just wondering what you like to see for temps on the IRFP MOSFETS. I had them running for about an hour and was seeing 40-42C. I think they had about 29V on them at the time.... I need to run the test again and also install the second channel and listen to it for a while. Then when I do get the case, it's just a mtter of plug and play.



As to the 4ohm tweaks, I think I will stay with 8ohm as this will go in the dedicated room should I really like them. If I don't or they don't fit well in my chain, I might consider going 4ohm for keeping them in the lab with the Tektons. Hugh has assured me that I will like them, so I guess all is well. (Kidding ya Hugh)


JT
 
AN

Thanks X


Hello V, the sink is 6x12 inches, for one channel. The other side has the power supply and their sinks, but they don't really need that sink. I could mount them elsewhere and run a flying lead over if needed,(one MOSFET on each sink) but I really don't want to do that unless I need because of lead length and it will pass through the contained transformer area. The power supply leads have to pass through that area, but they go through a copper pipe as they pass through and are shield by the pipe.


The object for the component layout was to isolate the AN and input from as much of the noisy stuff as possible.
 

Attachments

  • 20200211_061743.jpg
    20200211_061743.jpg
    197.9 KB · Views: 296
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
Neat setup - can you show us an oblique view so we can see the whole chassis front and heatsinks? Was it designed to have the clamshell concept of shielding the trafo in the middle? If you have a link where this chassis can be bought that might be of interest for others making this amp.

What is your bias current through the MOSFETs? A 6in x 12in is a tad smaller than my 4Ux300mm and my temps are 56C. So I am wondering how many watts are you dissipating into the sink as your temps are 10C lower.
 
Hi JT,

How is your transformer mounted?
If it is bolted, it looks like it could make an additional single turn with the copper pipe.
This bolt, pipe and chassis becomes a shorted winding over the secondaries and will overload/overheat the primary,
until something melts, hopefully the fuse.

Danny
 

Attachments

  • Toroid_mounting.gif
    Toroid_mounting.gif
    3.8 KB · Views: 427
You have very low temps on your MOSFETs - must be getting excellent cooling. Mine are about 68C and heatsink is 56C.

I have heard 55C is limit for burn safety. MOSFETs can handle up to 150C, but they melt after that.


I'm going to run that test again... there's noway we have that much diff, there has to be an error in the data.



P -Channel 4.9v on Gate 27.6V on Drain 380mV on Source
N-Channel 4.7V - 27.6V - 378mV
 
I checked the temps again, moving the probe left and right and all over. I'm seeing up to 52C so far. I had to shut down the test after about 45mins as I had take mom to the doctor.


@ Danny , it's bonded to one side sitting off the metal, there is no bolt, but thanks for the concern, I would rather have you say something than not. :)
 
Chassis

Here ya go X


The middle section has no holes in it. I milled either end, and smoothed it down, so you can snake wires both front and back and then drilled a hole through both where you see the copper pipe. 3/8 hole and pipe worked perfect.


The hotest point I see is right in the side of the MOSFETs where the have the indentation. I shoot the heatsink from the other side right where the chip is and I get about 48-50C, so the transfer is very good. Keep in mind this test was with the case sitting to open air. I might have to go ahead and run a longer leads and drop one of those FETs on the other sink. I'd have to drill another hole, as I don't want those in with the other wires I have going through now. The power supply don't hardly get warm, so I'm thinking I could mount them to the tunnel in the middle and give the FET full run of the sink. Then we would have one 6x12 sink for each and be able to go up on the voltage if desired, but mainly I think we would stay below 50C with no issues.



I decide tomorrow after I run a long idle test and check temps. I'll put the top cover and face on too and see where it will be in the real world one it leaves the bench.





The sellers' store is closed, so no link is available. I'd be happy to provide one when they open again.




Note: I have no Kapton tape.
 

Attachments

  • 20200211_072502.jpg
    20200211_072502.jpg
    493.9 KB · Views: 413
  • s-l400.jpg
    s-l400.jpg
    15.1 KB · Views: 430
Last edited:
Slots

Here's a pic of the slots I milled on the two verticals both front and back side: You can see the tranny outputs go through one... there is a spacer place that is in place there and holds the wires in place nicely. There are also zip ties on both sides with isolate the wires and stand them off the metal of the metal in all directions.
 

Attachments

  • slot.jpg
    slot.jpg
    256.6 KB · Views: 398
Last edited:
Founder of XSA-Labs
Joined 2012
Paid Member
For grins I connected my 5ohm PTT6.5 and RS28F TL’s to my 8ohm Alpha Nirvana today. Not expecting it to work well, but surprised that it sounds glorious. Not a problem at all driving this load:
868141d1597513796-simple-passive-harsch-xo-using-ptt6-5-rs28f-waveguide-ptt6-5-rs28f-wg300-xo-meas-dats-impedance-jpg


XO schematic:
868139d1597513796-simple-passive-harsch-xo-using-ptt6-5-rs28f-waveguide-ptt6-5-rs28f-wg300-xo-schematic-jpg


With Matytinman’s special recommendation of Miflex but in 0.1uF value.
869067d1597875523-simple-passive-harsch-xo-using-ptt6-5-rs28f-waveguide-ptt6-5-rs28f-wg300-xo-miflex-dayton-resistors-jpg


Here is the speaker:
856179d1593297126-simple-passive-harsch-xo-using-ptt6-5-rs28f-waveguide-cf790e3c-21c3-4918-bb37-81eac893bc89-jpeg


Really enjoying Diana Krall’s new release: “How Deep is the Ocean”. High DR recording, I am sure.