The Benchmark AHB2 amplifier uses a switch mode power supply and it has vanishingly low noise.
Hello
Once a wild there was hdd failure, I was fed-up of that. So now I use an Internal Solid State Drive (SSD), they cost more than standard hdd, so I use a smaller one, but I still do back-up on a usb-hdd.
Bye
Gaetan
Thanks, Bfpca. 🙂
I'm not sure I agree that 3mm brass rod in a matching hole on the front panel won't be able to cope with the bending weight of 2x 300VA traffos on the front corners (3kg each) - but your idea of a (bolted) rectangular steel plate across the front is an excellent idea! 🙂
(I don't need plywood as I use 6mm (1/4") Sorbothane pads under all my power traffos, to reduce transferred vibration to the case.)
Andy
Finally, thanks to the work slowdown (virus) I managed to finalize test build of Alpha Nirvana (8ohm).
One pic attached, more pictures at
http://www.slowbears.com/flickr/AlphaNirvana/
Did not test with music yet, just sinus and square waves at 1khz.
Looks good on the scope. Clipping (upper half) at approx 51Vpp.
All oscilloscope photos are at 1kHz, 8Ohm resistive load.
Now will build 2nd channel, and then chassis with 2 fans (see pic).
One thing I had to change - increase R5 to 11k.
Originally, Vout offset was -266mV, now (with R4=3k3, R5=11k) it's -20mV.
My speakers are 6 Ohm (JBL Studio 590), so I guess this build is something in between both published schematic (4 and 8 Ohm).
All my tests did not last long, as the output IRFPs were running at 100 degrees C.
They heat up very quickly, while heatsink and water bucket takes MINUTES to catch up...
Just for this test I used Keratherm pads, but I have a feeling they do not conduct heat very well.
I used input pair: BC556C (matched beta and Vbe), VAS: TTC004B, Q4: 2N5401C, output IRFP240/9240, Re: 0.15 Ohm
No snubbers, no zeners, no lifted ground (I always use 2 separate wires for input ground, and dirty ground and never had a need to use lifted ground).
From what I see, VAS device can easily be TO92 small transistor (E.g. 2N5551)
Once I have proper chassis and cooling, will do more tests.
One pic attached, more pictures at
http://www.slowbears.com/flickr/AlphaNirvana/
Did not test with music yet, just sinus and square waves at 1khz.
Looks good on the scope. Clipping (upper half) at approx 51Vpp.
All oscilloscope photos are at 1kHz, 8Ohm resistive load.
Now will build 2nd channel, and then chassis with 2 fans (see pic).
One thing I had to change - increase R5 to 11k.
Originally, Vout offset was -266mV, now (with R4=3k3, R5=11k) it's -20mV.
My speakers are 6 Ohm (JBL Studio 590), so I guess this build is something in between both published schematic (4 and 8 Ohm).
All my tests did not last long, as the output IRFPs were running at 100 degrees C.
They heat up very quickly, while heatsink and water bucket takes MINUTES to catch up...
Just for this test I used Keratherm pads, but I have a feeling they do not conduct heat very well.
I used input pair: BC556C (matched beta and Vbe), VAS: TTC004B, Q4: 2N5401C, output IRFP240/9240, Re: 0.15 Ohm
No snubbers, no zeners, no lifted ground (I always use 2 separate wires for input ground, and dirty ground and never had a need to use lifted ground).
From what I see, VAS device can easily be TO92 small transistor (E.g. 2N5551)
Once I have proper chassis and cooling, will do more tests.
Attachments
Nice work, Minek.
I have a SigLent CRO too!
If your offset is not correct it generally means your diff pair is not running equal current.
Is that your gall bladder on top??
HD
I have a SigLent CRO too!
If your offset is not correct it generally means your diff pair is not running equal current.
Is that your gall bladder on top??
HD
Thanks Hugh.
Even in sim it's extremely difficult to make AN LTP currents equal..
But the way it turned out, it's acceptable (for this topology) - current diff is within 4% and output Vdc is 20mV
Here are both boards completed. Slightly dense (8cm x 4cm)...
Working on the chassis now.
What is the exact purpose (beyond general 'protection') of diodes attached to emitters of LTP?
I can't imagine any situation when they would be conducting.
Attachments
I concede that the fb transistor probably does not need diode protection, but the input stage transistor should be protected at base/emitter with a diode. It is not mandatory, but it is in good practice.
HD
HD
Before I blow anything up...
...I need some advice.
In these hard DIY times I was thinking of powering my ANs with minimalistic dual Meanwell LRS-350-24s which can be boosted up to 28V.
I tested and these SMPSes have floating + and - terminals while the ground terminal is in contact with the metal case.
By construction, the metal cases will be grounded to the amp chassis.
Do you think the following schematics is safe? What about ground loops?
Jacques
...I need some advice.
In these hard DIY times I was thinking of powering my ANs with minimalistic dual Meanwell LRS-350-24s which can be boosted up to 28V.
I tested and these SMPSes have floating + and - terminals while the ground terminal is in contact with the metal case.
By construction, the metal cases will be grounded to the amp chassis.
Do you think the following schematics is safe? What about ground loops?
Jacques
Attachments
C'mon Jacques - sure money is in shorter supply now (due to Corona-virus restrictions) - but you've built a super sounding (but, admittedly, not inexpensive!) amp ... why would you want to degrade its performance by not using an SLB? 😕
Andy
The connection seems fine to me.
Just make absolutely sure they have floating + and - outputs, i.e. not connected to ground.
Hi JA,
Should be OK, but as Skylar says, be absolutely sure it's floating.
Try it before you connect to the amp, of course, with a small resistor from +Vcc to -Vcc.
I have had good experiences with Meanwell supplies, they are well engineered.
Cheers,
Hugh
Should be OK, but as Skylar says, be absolutely sure it's floating.
Try it before you connect to the amp, of course, with a small resistor from +Vcc to -Vcc.
I have had good experiences with Meanwell supplies, they are well engineered.
Cheers,
Hugh
Jacques Antoine,
If the Meanwells don’t add noise - it’s a worthy thing to try and use. Some SMPS I have used are absolutely devoid of mains hum at 60/120/180Hz. I mean -130dB noisefloor there as measured with my FFT. I have not used large Meanwells but have heard they perform well. Some brands like Connexelectronic are just awful noise injectors with both added hum and hiss that could not be filtered out.
14.6A at 24v is a beefy supply and you could run both channels easily.
You may want to put a CL-60 or 8D-20 NTC between the MW chassis Ground and your Chassis ground to keep the dirty ground from contaminating your clean 0v analog ground. So don’t attach the GND terminal directly to the mains protective earth - but through the NTC as a ground loop breaker.
Good luck!
If the Meanwells don’t add noise - it’s a worthy thing to try and use. Some SMPS I have used are absolutely devoid of mains hum at 60/120/180Hz. I mean -130dB noisefloor there as measured with my FFT. I have not used large Meanwells but have heard they perform well. Some brands like Connexelectronic are just awful noise injectors with both added hum and hiss that could not be filtered out.
14.6A at 24v is a beefy supply and you could run both channels easily.
You may want to put a CL-60 or 8D-20 NTC between the MW chassis Ground and your Chassis ground to keep the dirty ground from contaminating your clean 0v analog ground. So don’t attach the GND terminal directly to the mains protective earth - but through the NTC as a ground loop breaker.
Good luck!
Last edited:
Thank you all for the replies.
X: you mean that I should insulate the MW chassis from the metal amp chassis like mounting them on a wood panel or something ?
Then they could only be in electrical contact with the metal case through the NTCs?
X: you mean that I should insulate the MW chassis from the metal amp chassis like mounting them on a wood panel or something ?
Then they could only be in electrical contact with the metal case through the NTCs?
Hello,
I would like to congratulate and thank all of you who have participated in the creation of this fantastic class A amplifier ... and that it be shared in this great audio forum!
Being Alpha Nirvana 39w (27db gain approximately) the best candidate for the next construction, I am looking for a preamp that makes a good match. The weakest source signal I have, comes from the 550mV phono preamp output.
The highest signal is from the CD 2V ~ reader.
In the course of this thread, I seem to have read that the Alpha Nirvana's input sensitivity is 630mVRMS and this makes me doubt the choice of preamp gain to use.
I am thinking of Yarra / Melbourne adjusted to 12dB of gain, it would be fine for the 630mV of the pream phono but I am left over for 2Vout CD player.
Any comment will be welcome .
Thanks and best regards
I would like to congratulate and thank all of you who have participated in the creation of this fantastic class A amplifier ... and that it be shared in this great audio forum!
Being Alpha Nirvana 39w (27db gain approximately) the best candidate for the next construction, I am looking for a preamp that makes a good match. The weakest source signal I have, comes from the 550mV phono preamp output.
The highest signal is from the CD 2V ~ reader.
In the course of this thread, I seem to have read that the Alpha Nirvana's input sensitivity is 630mVRMS and this makes me doubt the choice of preamp gain to use.
I am thinking of Yarra / Melbourne adjusted to 12dB of gain, it would be fine for the 630mV of the pream phono but I am left over for 2Vout CD player.
Any comment will be welcome .
Thanks and best regards
I use the Aksa Lender preamp as it sounds great and is simple to build. It has multiple input selector switch and RCA jacks like the Yarra. The Yarra will work great too, just a much more involved build.
In general, a 12dB gain preamp is great for general purpose use. The difference between your CD player and phono might be handled with a simple resistive attenuator between the RCA input at the jack. I know, attenuate it just to amplify it. I assure you though, you won’t know the difference. 🙂
In general, a 12dB gain preamp is great for general purpose use. The difference between your CD player and phono might be handled with a simple resistive attenuator between the RCA input at the jack. I know, attenuate it just to amplify it. I assure you though, you won’t know the difference. 🙂