Aleph J Schematic
- Thread starter Nelson Pass
- Start date
Sorry to resurrect an old thread.
I built an Aleph J in 2012 using P Daniels boards. I compared it to my factory Aleph 3 around that time but preferred the Aleph 3 which is still in my main system. The amp has worked flawlessly and is the amp I use in my workshop with my old test speakers.
I am (finally!) building my M2 (using teabag boards I bought many years ago). I decided to re-use the case and psu from an F5 of even older vintage that I hardly use. Before taking the F5 boards out, I tested the F5 and was surprised how much clearer it sounds than my Aleph J. I thought the dark sound I was used to hearing was due to the cheap old speakers in my workshop, but the F5 proved me wrong. A few more A/B comparisons of well known tracks confirmed that the Aleph J sounds a lot darker than the F5. I put it down to the character of the amps.
Today I heard one channel of my new M2, and again was surprised how much clearer it sounds than the Aleph J.
If it were an Aleph vs F5/M2 comparison I may have blamed mosfets vs. jfets in the front end, but the J has jfets as well. So, I am none the wiser.
Note: All listening in my workshop was using the same single ended source with the Js IN- grounded.
I have checked all my notes and photos (I have attached an old photo) and hope someone can give me a hint what might be the issue.
PSU (for each rail):
Amp parts used are:
The 5pF capacitor is a ceramic one, marked "CM A 5.00 1KV" (I assume the one specified in the bom I downloaded at the time).
The 1uF capacitor I assume is a Wima MKS2 (red box cap, marked "1u K 63").
Is there anything in the above configuration that could explain the dark sound?
I have read that replacing the 5pF by a 10pF may have sonic benefits. I just received some 10pF mica from member rhthatcher (thanks!) and had planned to replace the 5pF ceramics with these sometime in the future anyway.
There is little space on the board, but I have some 1uF IC axial polypropylene caps and some teflon tubing to avoid shorts.
Besides the 5pF and 1uF capacitors I mentioned above, the only unknown is the 3W 0R47 resistors used.
Can the capacitors or the 3W resistors I used have the described effect on the sound?
Thanks.
I built an Aleph J in 2012 using P Daniels boards. I compared it to my factory Aleph 3 around that time but preferred the Aleph 3 which is still in my main system. The amp has worked flawlessly and is the amp I use in my workshop with my old test speakers.
I am (finally!) building my M2 (using teabag boards I bought many years ago). I decided to re-use the case and psu from an F5 of even older vintage that I hardly use. Before taking the F5 boards out, I tested the F5 and was surprised how much clearer it sounds than my Aleph J. I thought the dark sound I was used to hearing was due to the cheap old speakers in my workshop, but the F5 proved me wrong. A few more A/B comparisons of well known tracks confirmed that the Aleph J sounds a lot darker than the F5. I put it down to the character of the amps.
Today I heard one channel of my new M2, and again was surprised how much clearer it sounds than the Aleph J.
If it were an Aleph vs F5/M2 comparison I may have blamed mosfets vs. jfets in the front end, but the J has jfets as well. So, I am none the wiser.
Note: All listening in my workshop was using the same single ended source with the Js IN- grounded.
I have checked all my notes and photos (I have attached an old photo) and hope someone can give me a hint what might be the issue.
PSU (for each rail):
- 225VA Amplimo toroid (2 x 11V; in series)
- 33000uF/50V UCC U32 (big blue can)
- Dale 3W 2K2 bleeder resistor
- 4 x 0R47 Panasonic parallel
- 33000uF/50V UCC U32 (big blue can)
- 4.7uF polypropylene
Amp parts used are:
- Dale RN55 resistors
- 3W 0R47 resistors unknown brand (similar size and shape but not the usual Panasonic)
- Toshiba jfets, probably from Spencer (trusted source) where I bought most of them
- matched mosfets from Hannes (member h_a, I assume trusted source)
- 10uF and 220uF caps are Elna Silmic II (bought from Mouser or Digikey)
The 5pF capacitor is a ceramic one, marked "CM A 5.00 1KV" (I assume the one specified in the bom I downloaded at the time).
The 1uF capacitor I assume is a Wima MKS2 (red box cap, marked "1u K 63").
Is there anything in the above configuration that could explain the dark sound?
I have read that replacing the 5pF by a 10pF may have sonic benefits. I just received some 10pF mica from member rhthatcher (thanks!) and had planned to replace the 5pF ceramics with these sometime in the future anyway.
There is little space on the board, but I have some 1uF IC axial polypropylene caps and some teflon tubing to avoid shorts.
Besides the 5pF and 1uF capacitors I mentioned above, the only unknown is the 3W 0R47 resistors used.
Can the capacitors or the 3W resistors I used have the described effect on the sound?
Thanks.
Attachments
without actual THD Spectra measurements, hard to say
but:
- maybe it's just your preference to sound character? Aleph J is more than certainly 2nd dominant, while your F5 easily can end as 3rd dominant; actual setting possible if you have iteration with 3rd trimpot, and only with use of THD Spectra measurement setup
- PSU caps are having effect on final sound of an amp; what you have there are old caps; not saying they're necessary bad, but difference can be checked only by swapping them and then hearing if there is difference; if you have same caps in F5, ignore this remark
but:
- maybe it's just your preference to sound character? Aleph J is more than certainly 2nd dominant, while your F5 easily can end as 3rd dominant; actual setting possible if you have iteration with 3rd trimpot, and only with use of THD Spectra measurement setup
- PSU caps are having effect on final sound of an amp; what you have there are old caps; not saying they're necessary bad, but difference can be checked only by swapping them and then hearing if there is difference; if you have same caps in F5, ignore this remark
I understand they are all different, but the comparison with the F5 and M2 makes the difference so clear that I have the feeling there may be something wrong with the J. I can't compare it to another J, but my 3 sounds bright in comparison, which is not something I would expect given the mosfet front end compared to the jfets in the J.
The F5 is the old version, so no P3.
Yes, the caps are old. The F5 has newer (only just haha) Panasonic 15000uF/63V caps (8 in total). Can old caps lead to a lack of highs? Do they affect different frequencies differently as they age? The bass is fantastic, better than the F5 or M2.
If the caps can be the issue, it will be worth replacing the caps to hear the difference. I have none lying around, so can't do that straightaway.
PS I like 2nd harmonic and love my H2.
Now, look at this below... and think... how's this applicable to your amp...
This topic is quite extensive and once you start thinking about it... you may end up using a wide braided copper wire to serve as a common return point, for the local decoupling caps (at least for the darn JFETs, but the MOSFETs could use those caps, too). FW AMP PCBs are missing these tracks and capacitors... The decoupling caps that will do the job fine (for analog amplifier needs....) are 1uF electrolytic and 0.1uF PP film in parallel for the JFET current, and 47uF electrolytic and 0.1uf PP film for the MOSEF currents (same HF frequency spectrum as with the JFETs, but much higher current swings needed for MOSFETs... hence, a 47uF local decoupling cap here...)
... you'll also consider creating a better common point at your filter caps (I could not see this in the photo you provided??), and using a much thicker and shorter ground wire between that point, and the AMP PCB GND... which can help in particular if you introduce that JFETs local decoupling later on.
... and then you may accept the fact that V+ and V- wires are the extensions of the picture above and that they should be treated the same as the common return wires (e.g. short, thick).
The yellow caps you used will be effective at high frequencies only if:
- the common point (at the PS) exists and is clearly defined as a stable reference.
- the capacitor's wires are very short (hint: look at the "BEST" on the above screenshot). Also, the PS PCB track width (or wire thickness and length in your case with no PS PCB being used), are defacto extensions of those capacitor leads... (and vice versa)
- add a 0.1uF cap in parallel to what you already have
Here's another one I found.... that'll help you a lot.... treat your V+ and V- to the JFETs the same as the Vcc of that IC:
The result will be... mind-blowing.
Last edited:
Thanks for all your replies.
Yes, I have wondered about local decoupling because it is so common and essential in digital circuits. It will be interesting to read up on that, and perhaps see where I can try it out on a simple amp.
I did not want to take the psu out of my M2, just to check the sound. Too much work.
I have 8 new (but old) Panasonic T-HA 15000uF/63V left over in my cupboard. These are the same ones in my old F5 psu (reused for the M2), which both sounded much clearer than the J. I thought I would mount those caps on a dual rail psu board that Tea-bag sold me years ago to make a new CRC psu for the Aleph J, but the board has max 35mm cap diameter and the caps are 40mm. I will probably have to make a psu out of perfboard if I want to use them, and that will take a bit of time.
I really wanted to know if the sound I was hearing (and not hearing) could be due to the caps, so tried the following:
Add a 15000uF/63V T-HA to the last C on each rail (using 1m long, not really suitable testleads, but worth a try anyway). The result? Highs are back and the Aleph J had never sounded so good.
So, the problem is not the character to the amp. It is the caps in the psu!
Without going crazy (buying Mundorf Mlytic), what currently is a good option?
I need to buy new caps for my Aleph J (obviously) and also for my F3 (which is nearly complete). I had planned to use more or those 33000uF/50V Nippon Chemi Con I have in my cupboard, but after hearing the difference today I think those are going in the bin. The T-HA with 63V rating would be nice for the F3, but I am unsure of their age. They have datecode 9352C3. If that means 1993 I may as well forget about using them.
Yes, I have wondered about local decoupling because it is so common and essential in digital circuits. It will be interesting to read up on that, and perhaps see where I can try it out on a simple amp.
I did not want to take the psu out of my M2, just to check the sound. Too much work.
I have 8 new (but old) Panasonic T-HA 15000uF/63V left over in my cupboard. These are the same ones in my old F5 psu (reused for the M2), which both sounded much clearer than the J. I thought I would mount those caps on a dual rail psu board that Tea-bag sold me years ago to make a new CRC psu for the Aleph J, but the board has max 35mm cap diameter and the caps are 40mm. I will probably have to make a psu out of perfboard if I want to use them, and that will take a bit of time.
I really wanted to know if the sound I was hearing (and not hearing) could be due to the caps, so tried the following:
Add a 15000uF/63V T-HA to the last C on each rail (using 1m long, not really suitable testleads, but worth a try anyway). The result? Highs are back and the Aleph J had never sounded so good.
So, the problem is not the character to the amp. It is the caps in the psu!
Without going crazy (buying Mundorf Mlytic), what currently is a good option?
I need to buy new caps for my Aleph J (obviously) and also for my F3 (which is nearly complete). I had planned to use more or those 33000uF/50V Nippon Chemi Con I have in my cupboard, but after hearing the difference today I think those are going in the bin. The T-HA with 63V rating would be nice for the F3, but I am unsure of their age. They have datecode 9352C3. If that means 1993 I may as well forget about using them.
I have no objection using any of brand name ones, 33mF, 25 or 35Vdc
snap-in, bought either at TME, Digi, Mou
rumor has it, Samwha of recent years is same good as brand name ones
now, I don't care what ppl are thinking of that, slapping 50uF (and up) motor-run as final cap on rail - paid of big, whatever main caps are
snap-in, bought either at TME, Digi, Mou
rumor has it, Samwha of recent years is same good as brand name ones
now, I don't care what ppl are thinking of that, slapping 50uF (and up) motor-run as final cap on rail - paid of big, whatever main caps are
I have versions of those three amps. F5 is brighter than Aleph J by no small bit. it is also PP. The AlephJ is single ended, and is colored in comparison. But a beautiful coloration it is! F5 likely has more bass control, but alephJ's bass, while perhaps looser, does sound more "powerful" to me as well. M2. I also have the same boards you used. My F5 was Peter Daniel boards. I seem to keep coming back to AlephJ....
Russellc
Hi Russell,
I have a factory Aleph 3 in my main system. It has been in there for >20 years so I understand when you say you keep coming back to the Aleph 😎
My J sounds darker than the 3. I tried it in my main system years ago (replacing the 3) but was disappointed.
My recent experimentation with the psu caps indicate they are the issue. I am considering adding some low impedance caps to the final C to see if that helps.
I have a factory Aleph 3 in my main system. It has been in there for >20 years so I understand when you say you keep coming back to the Aleph 😎
My J sounds darker than the 3. I tried it in my main system years ago (replacing the 3) but was disappointed.
My recent experimentation with the psu caps indicate they are the issue. I am considering adding some low impedance caps to the final C to see if that helps.
Well, the problem was indeed the caps as the quick-and-dirty test with the 2 x 150000uF/63V NOS Panasonic T-HA showed previously.
I have added 8 x 5600uF United Chemi Con KYB (low impedance), mounted on perfboard, to the last C. I only had time to test for a few minutes, but even with a cold amp and no burn in time it is much clearer and the darkness is gone. I hope to have some time later in the week to let the amp warm up properly and listen longer.
I am reforming some of the NOS 33000uF/50V I have in my drawer to replace the ones currently in the amp. I can disconnect the new 8x5600uF caps to listen if the amp sounds good after reforming or not. If not I am afraid the 33000uFs will be for the recycling bin.
I have added 8 x 5600uF United Chemi Con KYB (low impedance), mounted on perfboard, to the last C. I only had time to test for a few minutes, but even with a cold amp and no burn in time it is much clearer and the darkness is gone. I hope to have some time later in the week to let the amp warm up properly and listen longer.
I am reforming some of the NOS 33000uF/50V I have in my drawer to replace the ones currently in the amp. I can disconnect the new 8x5600uF caps to listen if the amp sounds good after reforming or not. If not I am afraid the 33000uFs will be for the recycling bin.