It is running for a couple hours now, how fantastic is this already sounding!
I took care of matching, measuring and even modest scoping. I had to replace one mosfet even after matching.
I did not alter the bias, it is at 1,8A+ and 1,8A- L and R.
There is a CLC power supply.
There is just one issue, it is running hot. Very hot. outside 65°C, fets go over 85.
They are all the same, more or less.
I have no issue with that, but the fets over 85 is no good.
I will add more heatsink, but it will be 'different'.
Many thanks to this wonderful forum and offcourse the author Nelson Pass for the design.
I would paste an image, I have no URL of the image.
Greetings,
Dieter
I took care of matching, measuring and even modest scoping. I had to replace one mosfet even after matching.
I did not alter the bias, it is at 1,8A+ and 1,8A- L and R.
There is a CLC power supply.
There is just one issue, it is running hot. Very hot. outside 65°C, fets go over 85.
They are all the same, more or less.
I have no issue with that, but the fets over 85 is no good.
I will add more heatsink, but it will be 'different'.
Many thanks to this wonderful forum and offcourse the author Nelson Pass for the design.
I would paste an image, I have no URL of the image.
Greetings,
Dieter
use Babysitter in meantime, while sorting increase in heatsinking
regarding pics - it's always best to put them here, not through external host/url
of course, properly sized
plenty of solutions to pics shrinking; one of free programs is IrfanView
regarding direct posting - "post reply" button on lower left of page is opening extended view posting window, you'll find "manage attachment" button, too
regarding pics - it's always best to put them here, not through external host/url
of course, properly sized
plenty of solutions to pics shrinking; one of free programs is IrfanView
regarding direct posting - "post reply" button on lower left of page is opening extended view posting window, you'll find "manage attachment" button, too
Last edited:
Dieter, everyone loves pictures, or as Zen Mod calls them, porn.
The best way to add pictures is to upload them to diyAudio. Post by hitting "Go Advanced" for Reply. Then click on the "paper clip" icon which opens a "Manage Attachments" window. From there you can choose picture files from your computer to upload to diyAudio. There is a maximum file size per picture so you may need to resize to pictures prior to uploading.
I just found a thread explaining all of this:
How to attach images to your posts.
I see that Zen Mod has addressed this while I was typing.
The best way to add pictures is to upload them to diyAudio. Post by hitting "Go Advanced" for Reply. Then click on the "paper clip" icon which opens a "Manage Attachments" window. From there you can choose picture files from your computer to upload to diyAudio. There is a maximum file size per picture so you may need to resize to pictures prior to uploading.
I just found a thread explaining all of this:
How to attach images to your posts.
I see that Zen Mod has addressed this while I was typing.
Thank you mr pass.
I was thinking something like that:
If you light up your wood burning stove then you don’t touch it right? You know it’s hot.
Thanks for explaining, didn’t see that button the first time. 😉
Now it’s playing on the bigger speakers. May I repeat how astonishing this is sounding? Really very good.
I was thinking something like that:
If you light up your wood burning stove then you don’t touch it right? You know it’s hot.
Thanks for explaining, didn’t see that button the first time. 😉
Now it’s playing on the bigger speakers. May I repeat how astonishing this is sounding? Really very good.
Attachments
Last edited:
Finished
The aleph I made is considered finished.
I played around with the bias setting, trying to raise, but it stays at 1,7. Varying just slightly.
Dc offset starts at max 45mV, as soon as it’s warm it’s 2 and 10mV.
I added slow two stage turn-on and thermal protection at 75deg.
Also added heatsinking, it’s now still 70deg at the sides, fully closed up, although thermal resistance will better once def closed. The heatsinking had to be incorporated in the aluminium covering.
I m missing a button and something for in the middle, pure aestically. Otherwise considered finished, it sounds very good to my ears, unbelievably detailed and indeed very fast rhythmic amp, it Sounds kinds of faster pace with rhythmic music. Low extension it has not indeed. A bit tame in the low end, although precise in delivering. I dsp’ed just a little bit with Roon, my most listening source, and now it’s just perfect for me and still can’t believe it will replace the hefty dussun amp I had for years (which seems to be a levinson derivate..).
Many thanks to all folks with all the useful info, albeit in other threads ;-). And mr Nelson Pass for this beautiful sounding and do-able design.
The aleph I made is considered finished.
I played around with the bias setting, trying to raise, but it stays at 1,7. Varying just slightly.
Dc offset starts at max 45mV, as soon as it’s warm it’s 2 and 10mV.
I added slow two stage turn-on and thermal protection at 75deg.
Also added heatsinking, it’s now still 70deg at the sides, fully closed up, although thermal resistance will better once def closed. The heatsinking had to be incorporated in the aluminium covering.
I m missing a button and something for in the middle, pure aestically. Otherwise considered finished, it sounds very good to my ears, unbelievably detailed and indeed very fast rhythmic amp, it Sounds kinds of faster pace with rhythmic music. Low extension it has not indeed. A bit tame in the low end, although precise in delivering. I dsp’ed just a little bit with Roon, my most listening source, and now it’s just perfect for me and still can’t believe it will replace the hefty dussun amp I had for years (which seems to be a levinson derivate..).
Many thanks to all folks with all the useful info, albeit in other threads ;-). And mr Nelson Pass for this beautiful sounding and do-able design.
Attachments
I have an original Aleph 5 paired with Spendor SP9/1. "Acoustic" music sounds amazing; beautiful mid-range, detailed, neutral & palpable, no bloom on male vocals, sense of rhythm and pace is about the best I've heard. I have never grown tired of listening to the music (classical & jazz) and have NOT updated my system for twenty years. But, granted; not ideal on "head-banging rock-n-roll".
Probably makes no difference. A proper class A amp takes constant power from the wall. Some of that is diverted to the speaker. The rest is waste heat.
When I abused a class A amp on the bench with test-tones, I could force plate dissipation down from 39W (at idle) to 22W (17W to load), and feel the difference in heat.
In un-clipped speech/music work the long term average load power is less than 1/10th the maximum power. So 39W at idle could be 35W playing REAL LOUD. And there are (rare) days I do not want to play just below clipping.
"less impedance speakers" may also be incorrect, depending on amp design. Most modern design aims for a steady current. Lower impedance at the same current means more voltage wasted in the amp. (True, there are other ways to do it.)
Modern devices will stand a LOT of heat. As Sir Pass says, sometimes it is careless customers who limit the allowable rise. Modern Silicon can run so hot the solder on its legs melts and it keeps working (though not forever). Boiling water is warm enough that the Leakage specs can be exceeded but a conservative designer can allow for that, so spit-sizzle is not necessarily "too hot". Temperature *swing* is worse than steady-hot. BOOM-ba-BOOM music in a class B amp wobbles temperature frequently. 30 years ago we were striving for <50degC *rise* (i.e. 25C to 75C).
Thank you all, with a bit of delay. Meanwhile this amp is playing very well, allthough I updated the system and some questions rise:
I try asking them in my own thread before starting a new one.
so there is a DIY P1.7 added, and now recently some klipsch horn speakers.
First the good news:
The whole combo is playing astonishingly good (is there a better word?) with a balanced source into the p1.7, in his turn connected balanced to the aleph 5. I had some grounding hum which I resolved now, that is to say, in fully balanced operation. This is very very good, dead quiet and really lively and entertaining.
BUT
when I play the aleph5 in single-ended mode, I now have more and lower-freq hum, it s clearly a 50hz rumble. It is not that important, because absolutely none of that in balanced mode, which I use.
But this rises a grounding question for the pro s over here:
So when using single ended, one must put a jumper between the in- pin2 and gnd pin1. Pin1 is said to be connected to CHASSIS GND directly, not audio ground (There’s a 10ohm thermistor between them). So, when playing single ended, you effectively connect IN- to chassis ground. Is that correct? I believe that not to be correct. Of lesser interest, the single-ended rumble I encounter is probably because i shielded that se signal internally to chassis ground, picking up 50hz rumble into the amp pcb. The shield is naked and touches chassis everywhere. So that’s an error.
Question2 MORE IMPORTANT
Balanced operation requires pin1 to be grounded, as to effectively get rid of signals on the shield. Shielded cables always must have their gnd connection on one side only. So what’s common then? To ground it at the female or male Xlr sockets (in my case pre or power amp, but this is a general question)?
Then the PRACTICAL PROBLEM
Then, for whoever cares, the most irritating problem, probably concerning the P1.7. Balanced sources sound considerably better than unbalanced. I still can’t see a reason for it. So the same goes as above, only is the IN- of the preamp grounded this time. This makes effectively a connection between IN- and chassis GND, which to my understanding has to be seperated. The P1.7 has the characteristic to be able to output balanced while receiving unbalanced. So not only is there the theoretical 3db (or was it 6?) of less pre-amplification, it also sounds more dull, not as entertaining as with a balanced source. This has always been the case, but is much more portrayed now that I have high-sensitivity klipsch’s instead of the former unefficient but very very good b&w’s.
This problem stays true even when using another (unbalanced) amp, I have a spare Quad306.
So what would be the correct use of the little bridge in the Xlr connector, knowing that pin1 is (has to be?) chassis ground and not audio ground?
Just looking to get the absolute most out of it, as it is right now only with balanced sources (only my CD player has this, a XONO derative i am also making will join that). My most-used digital collection has only single-ended.
I try asking them in my own thread before starting a new one.
so there is a DIY P1.7 added, and now recently some klipsch horn speakers.
First the good news:
The whole combo is playing astonishingly good (is there a better word?) with a balanced source into the p1.7, in his turn connected balanced to the aleph 5. I had some grounding hum which I resolved now, that is to say, in fully balanced operation. This is very very good, dead quiet and really lively and entertaining.
BUT
when I play the aleph5 in single-ended mode, I now have more and lower-freq hum, it s clearly a 50hz rumble. It is not that important, because absolutely none of that in balanced mode, which I use.
But this rises a grounding question for the pro s over here:
So when using single ended, one must put a jumper between the in- pin2 and gnd pin1. Pin1 is said to be connected to CHASSIS GND directly, not audio ground (There’s a 10ohm thermistor between them). So, when playing single ended, you effectively connect IN- to chassis ground. Is that correct? I believe that not to be correct. Of lesser interest, the single-ended rumble I encounter is probably because i shielded that se signal internally to chassis ground, picking up 50hz rumble into the amp pcb. The shield is naked and touches chassis everywhere. So that’s an error.
Question2 MORE IMPORTANT
Balanced operation requires pin1 to be grounded, as to effectively get rid of signals on the shield. Shielded cables always must have their gnd connection on one side only. So what’s common then? To ground it at the female or male Xlr sockets (in my case pre or power amp, but this is a general question)?
Then the PRACTICAL PROBLEM
Then, for whoever cares, the most irritating problem, probably concerning the P1.7. Balanced sources sound considerably better than unbalanced. I still can’t see a reason for it. So the same goes as above, only is the IN- of the preamp grounded this time. This makes effectively a connection between IN- and chassis GND, which to my understanding has to be seperated. The P1.7 has the characteristic to be able to output balanced while receiving unbalanced. So not only is there the theoretical 3db (or was it 6?) of less pre-amplification, it also sounds more dull, not as entertaining as with a balanced source. This has always been the case, but is much more portrayed now that I have high-sensitivity klipsch’s instead of the former unefficient but very very good b&w’s.
This problem stays true even when using another (unbalanced) amp, I have a spare Quad306.
So what would be the correct use of the little bridge in the Xlr connector, knowing that pin1 is (has to be?) chassis ground and not audio ground?
Just looking to get the absolute most out of it, as it is right now only with balanced sources (only my CD player has this, a XONO derative i am also making will join that). My most-used digital collection has only single-ended.
First, let's be sure that we're all using the same pin numbering associated with their normal uses for the XLR connection. There are a few instances where your pin numbering doesn't seem to align. That could be a fundamental issue, or it could just be typos.
As an example, when using SE inputs on some amps, you would normally short pins 1 and 3 vs. pins 1 and 2. Pin 3 usually carries the inverted signal, not pin 2.
Depending on your wiring, that could be part of your issue. Usually the "hot" pin of the RCA for an SE input would be tied to pin 2 of the XLR. If you are actually shorting Pin 2 to ground with that wiring....
As an example, when using SE inputs on some amps, you would normally short pins 1 and 3 vs. pins 1 and 2. Pin 3 usually carries the inverted signal, not pin 2.
Depending on your wiring, that could be part of your issue. Usually the "hot" pin of the RCA for an SE input would be tied to pin 2 of the XLR. If you are actually shorting Pin 2 to ground with that wiring....
Attachments
That could well be the error, I thought it to be the other way around.First, let's be sure that we're all using the same pin numbering associated with their normal uses for the XLR connection. There are a few instances where your pin numbering doesn't seem to align. That could be a fundamental issue, or it could just be typos.
As an example, when using SE inputs on some amps, you would normally short pins 1 and 3 vs. pins 1 and 2. Pin 3 usually carries the inverted signal, not pin 2.
Depending on your wiring, that could be part of your issue. Usually the "hot" pin of the RCA for an SE input would be tied to pin 2 of the XLR. If you are actually shorting Pin 2 to ground with that wiring....
I have to check this in power and pre-amp.
Thank you for this insight!
EDIT
But still: you have to tie this pin to ground, that is, chassis ground? Effectively tieing a half of the amp to chassis ground while the audio ground is separated by a 10ohm resistor?
Thank you!
^ I intentionally avoided answering that question b/c it is often debated. However, I think it's safe to simply provide what I believe is done in all the First Watt amps and some of the Pass Labs amp schematics I've seen. I've never seen the schematic for a Aleph 5, but I've seen a few related schematics.
I would tie pin 1 to audio ground (wherever the signal return for your output is tied) vs. chassis/safety earth. In the situation where you're intentionally using a 10ohm resistor as a "ground lift", it fits with that use case... in my opinion. Others may have differing opinions, and theirs are just as valid.
I would tie pin 1 to audio ground (wherever the signal return for your output is tied) vs. chassis/safety earth. In the situation where you're intentionally using a 10ohm resistor as a "ground lift", it fits with that use case... in my opinion. Others may have differing opinions, and theirs are just as valid.
Not sure about that. I used for many years an over biased Aleph J and in my system and it performed well on rock n roll.
Your result could be due to something else.