Lesson learned.
Thanks AudioFreak I knew I was missing something, you are right all the way!!! 🙂
Thanks AudioFreak I knew I was missing something, you are right all the way!!! 🙂
Forgot to say that in addition to the inductors I also made the following additios to the Aleph 4.
Bypassed the CCS zener to the dif pair with a 0.1u and also added a 0.47u to each rail, all of these caps are Wima MKP 10.
All in all very worhtwhile mods.
Did some extensive listening yesterday night. As soon as the break-in was almost over (about 3 hours) everything came in very nicely. No problem with bass slam as I mentioned before.
Not to say that the sound is dry but it seems to me that some decay information is missing, think that this has to do with the low voltage I now have on the rails (32V down from 44V).
Bypassed the CCS zener to the dif pair with a 0.1u and also added a 0.47u to each rail, all of these caps are Wima MKP 10.
All in all very worhtwhile mods.
Did some extensive listening yesterday night. As soon as the break-in was almost over (about 3 hours) everything came in very nicely. No problem with bass slam as I mentioned before.
Not to say that the sound is dry but it seems to me that some decay information is missing, think that this has to do with the low voltage I now have on the rails (32V down from 44V).
CLCRC PSU
Now having the LCRC PSU I plan to add jet another section to it not because I need it but to get the rail voltage up again, And also because I want the inductors in that position to protect the caps which are only rated to 50 V (63V surge).
So the next step is to add some caps on the front to end with a CLCRC type of filter (kind of overdone???).
One question, is there a way to calculate the amount of capacitance needed at the front to get a desiered voltage?
Happy listening!!!
Now having the LCRC PSU I plan to add jet another section to it not because I need it but to get the rail voltage up again, And also because I want the inductors in that position to protect the caps which are only rated to 50 V (63V surge).
So the next step is to add some caps on the front to end with a CLCRC type of filter (kind of overdone???).
One question, is there a way to calculate the amount of capacitance needed at the front to get a desiered voltage?
Happy listening!!!
Re: CLCRC PSU
Your rail voltage is really a mathematical relation to your input voltage and the Recifier stage. Although adding an inductor pulls down the rail voltage, there is no way to boost it after rectification without affecting current flow.
Capacitors only store energy to release it on the next cycle, which gives you a clean DC rail and adds some energy (stored) to the circuit.
You can try using HEXFREDs instead of a bridge as they are more effcient, change your Transformer or insert a variac wired 0-140 VAC to boost your input voltage. As the transformer will work just as well with a slightly higher input about 5% you will get o higher output from it. This higher output means more energy for the Bridge to rectify. There is on caveat, that is less available current from the variac once you go above line votage to step up mode.
You will also need to make sure you get a 10 to 15 Amp variac for the job. It will need to stay in circuit all the time.
Anthony
apassgear said:Now having the LCRC PSU I plan to add jet another section to it not because I need it but to get the rail voltage up again, And also because I want the inductors in that position to protect the caps which are only rated to 50 V (63V surge).
So the next step is to add some caps on the front to end with a CLCRC type of filter (kind of overdone???).
One question, is there a way to calculate the amount of capacitance needed at the front to get a desiered voltage?
Happy listening!!!
Your rail voltage is really a mathematical relation to your input voltage and the Recifier stage. Although adding an inductor pulls down the rail voltage, there is no way to boost it after rectification without affecting current flow.
Capacitors only store energy to release it on the next cycle, which gives you a clean DC rail and adds some energy (stored) to the circuit.
You can try using HEXFREDs instead of a bridge as they are more effcient, change your Transformer or insert a variac wired 0-140 VAC to boost your input voltage. As the transformer will work just as well with a slightly higher input about 5% you will get o higher output from it. This higher output means more energy for the Bridge to rectify. There is on caveat, that is less available current from the variac once you go above line votage to step up mode.
You will also need to make sure you get a 10 to 15 Amp variac for the job. It will need to stay in circuit all the time.
Anthony
There is a great free program for designing power supplies here:
http://www.duncanamps.com/psud2/index.html
http://www.duncanamps.com/psud2/index.html
Re: CLCRC PSU
If you keep the capacitance directly after the bridge the same as it was initially, you'll end up with pretty close to the original output voltage less the loss from the DCR of the inductor.
apassgear said:Now having the LCRC PSU I plan to add jet another section to it not because I need it but to get the rail voltage up again, And also because I want the inductors in that position to protect the caps which are only rated to 50 V (63V surge).
So the next step is to add some caps on the front to end with a CLCRC type of filter (kind of overdone???).
One question, is there a way to calculate the amount of capacitance needed at the front to get a desiered voltage?
Happy listening!!!
If you keep the capacitance directly after the bridge the same as it was initially, you'll end up with pretty close to the original output voltage less the loss from the DCR of the inductor.
Power Supply design
There is a very good discussion of the elements of solid state power supply design in the Radio Amateurs' Handbook. You should be able to find a copy of it in nearly any library. It is published every year. A good chapter on solid state power supplies has been included in every issue for the last 20 years. I have recently checked out the issue from 1981 and found that it included a good explaination of the effect both capacitence and choke input configuration has on all of the other components, output voltage, and noise. It also includes all of the design formulias need to do a class A design engineering job.
The text is written for the amateur electronics DIY'r.
PS; I don't think the lower voltage is causing you a problem. I suspect that the output is simply much cleaner then your are used to hearing and, therefore sounds thiner then it did before.
In any case, good luck with your project.
Tom
There is a very good discussion of the elements of solid state power supply design in the Radio Amateurs' Handbook. You should be able to find a copy of it in nearly any library. It is published every year. A good chapter on solid state power supplies has been included in every issue for the last 20 years. I have recently checked out the issue from 1981 and found that it included a good explaination of the effect both capacitence and choke input configuration has on all of the other components, output voltage, and noise. It also includes all of the design formulias need to do a class A design engineering job.
The text is written for the amateur electronics DIY'r.
PS; I don't think the lower voltage is causing you a problem. I suspect that the output is simply much cleaner then your are used to hearing and, therefore sounds thiner then it did before.
In any case, good luck with your project.
Tom
Re: Power Supply design
Tom,
Thanks for the input, I have a 1957 edition of the Radio Handbook (Spanish edition) which I just saw having a lot of information an the subject. This book has been neglected always since I also have the RDH last edition but havent looked either on the subject since I need some time to take a dip.
I guess your comment is right-on on having a much cleaner sounding amp, I have thought of this as having reduced considerably the IMD.
After the addition of the inductors it was as having a different and much better amplifier.
After this experiance, in my view PSU's for SS (and for that matter also vacumm) is a much neglected issue. Of what I'm sure is that I'll never do again a PSU without the use of inductors for which now I have several at hand.
ThomasLMcLean said:
Tom,
Thanks for the input, I have a 1957 edition of the Radio Handbook (Spanish edition) which I just saw having a lot of information an the subject. This book has been neglected always since I also have the RDH last edition but havent looked either on the subject since I need some time to take a dip.
I guess your comment is right-on on having a much cleaner sounding amp, I have thought of this as having reduced considerably the IMD.
After the addition of the inductors it was as having a different and much better amplifier.
After this experiance, in my view PSU's for SS (and for that matter also vacumm) is a much neglected issue. Of what I'm sure is that I'll never do again a PSU without the use of inductors for which now I have several at hand.
Attachments
Mark A. Gulbrandsen said:
Mark A. Gulbrandsen said:
Hi Mark,
I am still a bit confused here:
I have done the modification that WuffWaff did with his Aleph 5;
He wrote: "Since the last reports I changed the trimmers for R19 and R21 with resistors (49k and 390R) so now it runs at 2.4A bias and about 63% ac current gain (about 6.7A peak)."
Here is the thread: http://www.diyaudio.com/forums/showthread.php?postid=77253#post77253
So all in all he changed R19 from 221 K in the original scheme down to 49 K while also changing R21 to 390R.
This should have made for a higher Bias, or am I wrong?
Now Mark suggests leaving R19 totally out, which makes R19 getting to an infinitely higher resistance, which should make the bias lower....????????
How can this be explained?
Hopefully Mark and WuffWaff can help me out on this....
Mark, to what total bias does you Aleph 5 go when leaving out R19 altogether? How much dissipation does this bring per channel / per mosfet? Can you describe the change in sound?
All help appreciated...
Regards,
Lucas
Hello Lucas,
leaving R19 out will up the bias. Without it I get around 2.8A of bias.
R19 delivers a small current to the base of Q5. The bigger this current (R19 lower) the less the current through R20 has to be and the lower the value over R40-42. This way the bias is reduced.
hope this helps,
William
leaving R19 out will up the bias. Without it I get around 2.8A of bias.
R19 delivers a small current to the base of Q5. The bigger this current (R19 lower) the less the current through R20 has to be and the lower the value over R40-42. This way the bias is reduced.
hope this helps,
William
Thanks William,
This helps a lot!
Since my heatsinks do not get warmer than 47 degrees Celsius, I think I will soon try my Aleph 5 without R19.
Should I also change R21 then, or can I leave it at 390R?
Regards,
Lucas.
This helps a lot!
Since my heatsinks do not get warmer than 47 degrees Celsius, I think I will soon try my Aleph 5 without R19.
Should I also change R21 then, or can I leave it at 390R?
Regards,
Lucas.
Lucas,
everytime you change the bias you have to change R21 too, otherwise the ac-current-gain changes. (It gets lower if you only up the bias)
William
everytime you change the bias you have to change R21 too, otherwise the ac-current-gain changes. (It gets lower if you only up the bias)
William
wuffwaff said:
Sorry,
this is not quite correct!
Bigger current thru R19 requires BIGGER current thru R20 thus
less voltage drop on source resistors!
Those currents are a function of having 0.65 - 0.7 volts at the
base of the npn transistor!
wuffwaff said:
Again sorry,
this is also not correct as you multiply the ac gain with bias.
AC gain keeps constant if you change only R19! The only thing
besides bias to be changed is maximum current ability as this
is a function of bias and ac gain!
Uli
Actually I have left R19 out of my 2's and never messed with changing R21 at all. They sound great!!!!
Mark
Mark
Step n -- improve inductors
You can significantly improve el-cheapo inductors by adding a couple of very small inductors in series, for example one on each side. What this does is to reduce the series capacitance. Cost is negligible.
Another way to improve inductors is to use several in series.
I am one of those who like to place inductance on both ground AND hot lines (but you have to make the actual ground connections later in that case).
Petter
You can significantly improve el-cheapo inductors by adding a couple of very small inductors in series, for example one on each side. What this does is to reduce the series capacitance. Cost is negligible.
Another way to improve inductors is to use several in series.
I am one of those who like to place inductance on both ground AND hot lines (but you have to make the actual ground connections later in that case).
Petter
I am confused.
I remember that once the "Jaws" speaker shouted at me into the right understanding about the bias. 🙂
Reference: Aleph 5
My understanding is as follows:
R19 and R40-42 are related to the bias current. If R19 is increased, the bias current across R40-42 is increased. For example, If R19 is increased infinite (i.e. removed), the bias current will be increased from the original 1.5A close to 2A.
Meanwhile, R20, R21, R22-25, and R40-42 are related to the current gain. If R20 is increased (others remain the same), the AC current gain goes up. Or, if R21 is reduced (others remain the same), the AC current gain goes up.
JH
I remember that once the "Jaws" speaker shouted at me into the right understanding about the bias. 🙂
Reference: Aleph 5
My understanding is as follows:
R19 and R40-42 are related to the bias current. If R19 is increased, the bias current across R40-42 is increased. For example, If R19 is increased infinite (i.e. removed), the bias current will be increased from the original 1.5A close to 2A.
Meanwhile, R20, R21, R22-25, and R40-42 are related to the current gain. If R20 is increased (others remain the same), the AC current gain goes up. Or, if R21 is reduced (others remain the same), the AC current gain goes up.
JH
Re: Step n -- improve inductors
Petter,
Good idea to add a inductor to ground from the transformer.
But I would not be so sure to use El-cheapo chokes. One very important issue is to have a very low DCR and the second is to have no saturation for the current passed through it.
When you factor both of this requirements here is no way you can use el-cheapo inductors or air cores if you want some decent inductance, say 5 mH or more.
So on my book only a good gapped iron will do. Yes, dificult to get.
Petter said:
Petter,
Good idea to add a inductor to ground from the transformer.
But I would not be so sure to use El-cheapo chokes. One very important issue is to have a very low DCR and the second is to have no saturation for the current passed through it.
When you factor both of this requirements here is no way you can use el-cheapo inductors or air cores if you want some decent inductance, say 5 mH or more.
So on my book only a good gapped iron will do. Yes, dificult to get.
Hi Uli,
sorry for the late answer but I was away for a few days.
Of course you´re right with the first part since Kirchhoff´s law still applies here. (Must have been the wine and the time of day.........)
The second part comes from a measurement I made when I changed R19 to up the bias from 2 to 2.4A. Ac-current-gain changed from 63% to around 47%.
The only way I can explain this is that the current through the base of Q5 is higher and the current changes through R21 and the cap are still the same. Relatively this means that the AC part through Q5 gets less and so the ac-current-gain gets less.
william
sorry for the late answer but I was away for a few days.
Of course you´re right with the first part since Kirchhoff´s law still applies here. (Must have been the wine and the time of day.........)
The second part comes from a measurement I made when I changed R19 to up the bias from 2 to 2.4A. Ac-current-gain changed from 63% to around 47%.
The only way I can explain this is that the current through the base of Q5 is higher and the current changes through R21 and the cap are still the same. Relatively this means that the AC part through Q5 gets less and so the ac-current-gain gets less.
william
- Status
- Not open for further replies.