| bbakota2000 |
Hi guys,
Last couple of days I've been thinking how to make my Aleph 5 faster and more dynamic. I have one suggestion for the PSU, so you let me know what you think about it. It will have more filtration, more speed, very low output Z,... Sikorels and Fischer & Tausche elcos are really fast so that would be my choice.
Step up from this would be fast rectifier diodes (ST Micro BYW99W200 as Nelson suggested in one old thread).
What do you think? Will I be satisfied with these changes?
Merry Christmas everybody |
|
|
| analog_sa |
| Uh-oh! Another unauthorised use of Carlos snubber :) |
|
|
| bbakota2000 |
| What is Carlos snubber? |
|
|
| uli |
| quote: | Originally posted by analog_sa
Uh-oh! Another unauthorised use of Carlos snubber :) |
*ROFL*
Uli
:nod: :nod: :nod: |
|
|
| Panelhead |
| quote: | Originally posted by bbakota2000
Hi guys,
Last couple of days I've been thinking how to make my Aleph 5 faster and more dynamic. I have one suggestion for the PSU, so you let me know what you think about it. It will have more filtration, more speed, very low output Z,... Sikorels and Fischer & Tausche elcos are really fast so that would be my choice.
Step up from this would be fast rectifier diodes (ST Micro BYW99W200 as Nelson suggested in one old thread).
What do you think? Will I be satisfied with these changes?
Merry Christmas everybody |
This looks like a very nice power supply filter. Two things come to mind.
The Aleph's do not seem to respond to tweaking as much as most other designs. Using very high quality parts has much smaller effect on the sound. The better power supply may only provide a marginal improvement in sound quality.
I never use a single stage power supply filter anymore. If there is room a CLC is used, if not a CRC. All the bypass film caps are installed on the second stage of the filter. This takes care of the ringing. Using a bypass cap across a filter cap directly across from the diodes has always added noise, not remove it.
One easy tweak worth trying is to strap a fairly large (1- 2.2 ufd) film cap aross the +/- rail entry point on the circuit board. This buffered by the lead wire inductance and resistance. The sound quality should improve just as much as using the filter circuit drawn up.
For this application the Seimens stacked film and foil caps work great. The 1 ufd 100 volt are 25 cents and the 2.2 ufd 100v are a dollar at a local suplus shop. These have almost no sonic signature and have low inductance along with large bandwidth.
George |
|
|
| Mark A. Gulbrandsen |
You might wwant to consider converting them into a pair of 2's and use a Rosenblit grounded grid line stage to drive them. I've been running an over biased pair of 2's for over two years now(with zero problems!) and recently added the Grounded Grid in place of my Aleph L line stage. I was absolutely stunned with the addition of the grounded grid at just how fast and dynamic the 2's are, and its no doubt the best sounding my system has ever been. Its an amazing combnation of line stage and power amp to say the least. My 2's have a CLC supply running about 44 volts on the rails plus or minus a volt or so depending on the local power company. BUT.... I don't really think you will hear all that much difference with your power supply changes!!
If you do build the GG line stage be sure to make the feedback resistor 150K and use decent quality 5 uf coupling caps in place of the 2uf. I think I used Multicap PPMFX in mine and bypassed them with a .1 wonder cap. The Grounded Grid cost me less to build that it will cost for you to make those power supply changes!
AWSOME!
Mark |
|
|
| stadams |
bbakota2000,
Are your value selections arbitrary, following proven designs, or have you made calculations for these particular values? How did you determine the values for R1 and R2?
I like the idea of "upgrading" components in an attempt to bolster the performance of the power supply and the correct Sikorel capacitors would definitely be one of my choices. The R's that you have selected may provide you with the proper damping, however you really need to look at the power supply rails on an oscilloscope to see how the resistor/capacitor values that you have selected are effecting HF/RF noise on the rails. If you do use the ST diodes, noise may present itself to be less of a problem, but it is still worth looking at to really clean up the supply.
There are some other threads that you may want to look through when considering your design changes. At least one of the changes that you are considering could actually introduce noise problems if you do not look at the "snubbing" values experimentally. The ST diodes can help with this problem however. Also, do not forget about the effects the transformer has in the selection of the snubber components. But, if you are able to determine the noise effects experimentally, then the effect of the transformer is indirectly compensated and requires no direct attention.
Have you considered using regulation in your supply redesign?
Merry Christmas to you, and good luck with your redesign, |
|
|
| Mark A. Gulbrandsen |
Sorry guys BUT.....I would normally never discourage anyone I doubt he will hear ANY difference with those power supply mods. The only mod that might help would be a CLC type supply to get the noise floor lower than it is..... but then that also requires careful selection of the input devices to get them as quiet as possible and take advantage of a very low ripple supply.
Mark |
|
|
| jwb |
CRC power supply, or even a CRCRC supply, will provide an extremely quiet rail, and will not be suceptible to tank oscillation, which I think you might get with the original schematic, depending on the equivalent L and R of the capacitors specified.
The best way to keep ringing off your supply rails is to never have ringing in the first place. Use Schottky rectifiers if you can find ones big enough. Use soft/fast recovery diodes (e.g. Intl. Rectifier HFA series) otherwise. In my Aleph-X I used Schottky and do not regret it. I was initially afraid of the leakage power consumption, but these TO-247-packaged Schottky diodes are stone cold all the time, so no worries there.
If you DO use bad diodes, and you get ringing power supplies, you might try to find the article out there on the Internet, the title of which includes "Calculating Optimum Snubbers", if I recall correctly. |
|
|
| Mark A. Gulbrandsen |
"You might want to read NP's response to an old post of mine."
Yes, I remember that thread well. I even made a post if I recall. While there are alot of things that can be done or changed I think one needs to see most of them through to realize any noticable audible change except for one.... overbiasing the output stage has more positive effect on the quality of the sound from an Aleph than all the others on the list combined. Over time I've added such things as a CLC power supply, higher biasing, heavier wiring, etc and the higher biasing was the only one that had an immmediately apparent benefit and it was by a large margin.
The second most apparent change for the good was the addition of the Grounded Grid line stage. My transient response which I thought was good is now lightning fast.... really unbelievable!
A number of friends that frequently stop by and are familiar with how my system sounds also noticed the huge improvement.
Mark |
|
|
| jacco vermeulen |
| quote: | Originally posted by Panelhead
For this application the Seimens stacked film and foil caps work great. |
George,
which caps do you mean ?
Color/Number ? |
|
|
| bbakota2000 |
Mark,
Where can I find more info about the Rosenblit GG line stage? It's an intersting idea, I'll definitely try it. My Aleph 5 is also over biased, and I found it is working more "alive" with a tube preamp, but there were some other issues I didn't like. Now I'm having Aleph P as a preamp.
One more thing. You are completely right about overbiasing. It has enormous benefit compared to other mods (better caps, cap over zener diode, nice bypass cap in the PSU,..)
stadams,
I didn't do any calculations, so your idea is very good. Unfortunately I don't have an oscilloscope, but I'm trying to find one (as cheap as possible) for a long time now. I guess those things can't be cheap :bawling:
If the R value is as problematic as you suggest, it won't be a problem for me to experiment with them for a while. How can I calculate the correct R value?
I'll buy RFB 02 rectifier boards from peranders site. They can be used with or without snubbing caps, and they accept virtually any cap, so I'll experiment. |
|
|
| Mark A. Gulbrandsen |
The schematic is in the book "Audio Reality". This is Bruce Rosenblit's book. You might try writing to him.
Mark |
|
|
| stadams |
| quote: | Originally posted by bbakota2000
stadams,
I didn't do any calculations, so your idea is very good.
If the R value is as problematic as you suggest, it won't be a problem for me to experiment with them for a while. How can I calculate the correct R value?
|
bbakota2000,
The analysis of the power supply can be a somewhat lengthy process to explain so I will direct you to the article that jwb mentioned in his previous post:
Calculating optimum snubber values.
It is an excellent paper and has been referenced around diyaudio.com for a few years. Be forewarned that you will either have to measure or obtain the transformer's series resistance, self-inductance, and interwinding capacitance to make accurate calculations using this article.
Hurry up and get that O-scope!!
Good Luck, |
|
|
|
