CLS,
Is your 6C45 a tube or solid state rectified supply? Reguardless of what it is!
Take a look at this
Notice the cap across the choke! This person is after a lower ACR after getting rid of his AC ripple this is a very good idea IMO there are others I've seen do this as well! This should add some transparency to! And I do not like 4007 diodes I use NTE5817's very heavy duty but I like their sound notice the DCR of the choke just 26 ohm's good huh! Not low enough for me! I would be using a 17 lb 1 henry that has a DCR of 5 ohms!!! if I was interested or had to use a choke I may have to for my 833 amp the tube will use some 300 ma!
My iron is Hammond and I'm very satisfied with it especially my outputs which I would use for a 300B or 2A3 project even though they are over kill and you'd think the bandwidth would be poor buts its not I found this large iron has an ease at sound and great expansive space! no 6 or 10 pounders for me!
The noise differance you measured of 1.3 mv to 1.5 mv is just 1 1/4 db increase in noise and hardly audiable. I design for a noise level that I can tolorate at idle and that I can not hear thru the music this me! As I said there is a cost for over killing here!
The supply path that you are using for your b+ maybe ultra quite but it is not acceptable
for me it may be for your needs
but if I had to live with it in my amps they would suffer in to much of a dynamic loss 
and I would be 
I'm sure the iron is very pretty though What I'm advocateing is not conventional at all but it it makes my amps so dynamic that people hardly actually believe they are listening to just 10 watts if you care to try this I think you will be surprised at the sheer dynamics a SET 300B can deliever your friends will be shocked as my local club members are 
Heres what I do for my Monos Which run on about 140 ma each at 500 VDC so if you do this it would be an even lower noise floor my speakers are a highly modified Klipsch Forte at 98 db/watt I runn no preamp just CD direct from a Sony X77ES players variable outputs the preamp just gets in the way amp sensitivty is about 700 mv for full out most of the time my volume control is at 50% or 12 o'clock which can yield 90 to 100db at my chair listening room is 14 ft X 18 ft speakers on my long wall about 2 1/2 ft out I might add the sound stage is just huge on some recordings wider than my room and very deep I atribute some of the performance to my supplies and of course the output transformers which are real sleepers I also might add that I ground the positive output winding of my 28lb output transformers and float the negative why! because it sounds better everything got bigger anyway here it is one mono!
From a GZ37 only or equivlent slow turn on rectifier
----------------------------------------------D 20 uf solen polyprop
GZ37 C(600 uf) D (600 uf) D (600 uf)
----------------------------------------------D 12 uf solen polyprop
So both diode anodes are connected to the last 600 uf cap
I had used 2 1200 uf 350 volt caps stacked to get the 600 uf rating and used bleeder resistors which I had on hand to work with the 500 volt supply voltage I also stacked a pair of 40 uf solens to get the proper working voltage the other caps you see on my amps are for the DC filaments diode filtered as well but I plan on using a high current choke for this a common mode one would work better here but I don't have the voltage to work with!
You could use other tube rectifiers like the 5R4 but they would have to be choke input loaded and the b+ would fall as a result so its the GZ or a 5AR4 nos because new stock apparently does not meet slow turn on specs I've been looking at Russian damping tubes that have a 30 second turn on delay! And are very rugged and cheap!
If you use my configuration and have to turn it off wait at least 2 minutes or more for the GZ to cool done or risk arc over
As for the filament and diode connection with 2 diodes common the cathodes and connect each anode to a filament leg if your running DC if AC the use of the center tap is fine often the AC sounds better I have to use DC and I use the diodes in this manner the results of which are stunning and of course my cathodes are grounded if you have AC and no center tap I would use the diodes in this manner over resistors if you use resistors I would use bypass caps. Even if there was an imbalance between the legs I would not give this up! the dynamics of this went to another level when this was done enough to move pictures on my nieghbors walls her living room was 30 ft away seperated by garages between us!!!
Are yours DC or AC?
If you do try this use a supply that is current proportional to the one above for your driver tubes.
So as you can see! This is a extreamly fast power supply in reguards to both DCR and ACR with a total energy of 400 joules for both amps combined!!!
My feeling is that if you want Extoridinary Performance you have to use Extoridinary Concepts
SET12
Is your 6C45 a tube or solid state rectified supply? Reguardless of what it is!
Take a look at this
An externally hosted image should be here but it was not working when we last tested it.
Notice the cap across the choke! This person is after a lower ACR after getting rid of his AC ripple this is a very good idea IMO there are others I've seen do this as well! This should add some transparency to! And I do not like 4007 diodes I use NTE5817's very heavy duty but I like their sound notice the DCR of the choke just 26 ohm's good huh! Not low enough for me! I would be using a 17 lb 1 henry that has a DCR of 5 ohms!!! if I was interested or had to use a choke I may have to for my 833 amp the tube will use some 300 ma!
My iron is Hammond and I'm very satisfied with it especially my outputs which I would use for a 300B or 2A3 project even though they are over kill and you'd think the bandwidth would be poor buts its not I found this large iron has an ease at sound and great expansive space! no 6 or 10 pounders for me!
The noise differance you measured of 1.3 mv to 1.5 mv is just 1 1/4 db increase in noise and hardly audiable. I design for a noise level that I can tolorate at idle and that I can not hear thru the music this me! As I said there is a cost for over killing here!
The supply path that you are using for your b+ maybe ultra quite but it is not acceptable
for me it may be for your needs but if I had to live with it in my amps they would suffer in to much of a dynamic loss and I would be I'm sure the iron is very pretty though
What I'm advocateing is not conventional at all but it it makes my amps so dynamic that people hardly actually believe they are listening to just 10 watts if you care to try this I think you will be surprised at the sheer dynamics a SET 300B can deliever your friends will be shocked as my local club members are Heres what I do for my Monos
Which run on about 140 ma each at 500 VDC so if you do this it would be an even lower noise floor my speakers are a highly modified Klipsch Forte at 98 db/watt I runn no preamp just CD direct from a Sony X77ES players variable outputs the preamp just gets in the way amp sensitivty is about 700 mv for full out most of the time my volume control is at 50% or 12 o'clock which can yield 90 to 100db at my chair listening room is 14 ft X 18 ft speakers on my long wall about 2 1/2 ft out I might add the sound stage is just huge on some recordings wider than my room and very deep I atribute some of the performance to my supplies and of course the output transformers which are real sleepers I also might add that I ground the positive output winding of my 28lb output transformers and float the negative why! because it sounds better everything got bigger anyway here it is one mono! From a GZ37 only or equivlent slow turn on rectifier
----------------------------------------------D 20 uf solen polyprop
GZ37 C(600 uf) D (600 uf) D (600 uf)
----------------------------------------------D 12 uf solen polyprop
So both diode anodes are connected to the last 600 uf cap
I had used 2 1200 uf 350 volt caps stacked to get the 600 uf rating and used bleeder resistors which I had on hand to work with the 500 volt supply voltage I also stacked a pair of 40 uf solens to get the proper working voltage the other caps you see on my amps are for the DC filaments diode filtered as well but I plan on using a high current choke for this a common mode one would work better here but I don't have the voltage to work with!
You could use other tube rectifiers like the 5R4 but they would have to be choke input loaded and the b+ would fall as a result so its the GZ or a 5AR4 nos because new stock apparently does not meet slow turn on specs I've been looking at Russian damping tubes that have a 30 second turn on delay! And are very rugged and cheap!
If you use my configuration and have to turn it off wait at least 2 minutes or more for the GZ to cool done or risk arc over
As for the filament and diode connection with 2 diodes common the cathodes and connect each anode to a filament leg if your running DC if AC the use of the center tap is fine often the AC sounds better I have to use DC and I use the diodes in this manner the results of which are stunning and of course my cathodes are grounded if you have AC and no center tap I would use the diodes in this manner over resistors if you use resistors I would use bypass caps. Even if there was an imbalance between the legs I would not give this up! the dynamics of this went to another level when this was done enough to move pictures on my nieghbors walls her living room was 30 ft away seperated by garages between us!!!
Are yours DC or AC?
If you do try this use a supply that is current proportional to the one above for your driver tubes.
So as you can see! This is a extreamly fast power supply in reguards to both DCR and ACR with a total energy of 400 joules for both amps combined!!!
My feeling is that if you want Extoridinary Performance you have to use Extoridinary Concepts
SET12
Again, thanks a lot for all the tips
I'm afraid any larger energy storage can not be done to my amp
It's a stereo & the space is used up. Maybe I'll try it in my next amp.
About the cap across the choke, it looks a resonator to me. How do I determine the values? Is there any formula?
I'll try some diodes on cathodes (I'm using DC on filament BTW), and will update my progress.
I'm afraid any larger energy storage can not be done to my amp
It's a stereo & the space is used up. Maybe I'll try it in my next amp.
About the cap across the choke, it looks a resonator to me. How do I determine the values? Is there any formula?
I'll try some diodes on cathodes (I'm using DC on filament BTW), and will update my progress.
CLS said:About the cap across the choke, it looks a resonator to me. How do I determine the values? Is there any formula?
I'll try some diodes on cathodes (I'm using DC on filament BTW), and will update my progress.
It is a parallel resonant trap. Figuring out values for this isn't difficult at all. The resonant frequency is given by: w= (LC)^-0.5. In this case: w= [(2.6)(1.1E-6)]^-0.5
w= 591.31 rad/sec, which is: f= 94.11Hz, so he designed it for operation off 50Hz mains.
Q= Xl/Rs, so Q= [(591.31 * 2.6)]/26
Q= 59.1
Rp= Rs(1 + Q^2)
Rp= 26(1 + 59.1^2)= 90.8K
This gives the equivalent series resistance at 100Hz. Using that with the voltage divider formula will give the resultant AC ripple voltage. It will be quite low indeed. (in this case, it looks like some -75db(v) of attenuation) It's a neat trick if it works, but there is a potential problem with it: the AC circulating currents inside the LC trap are going to be very high, and may saturate that choke. That'll make it lose effectiveness. There is also a potential for that trap to contribute to low frequency instability.
Neat trick if it works.
Thanks
I also traced back the original article of that diagram and figured it out myself just now. Well, sort of.
So, I also saw the variation in the famous Opera M500 Modification that the tuning capacitor is in series with a 100R resistor. I suppose that is for tuning the disired Q of the resonator?
Then, again, can we use a diode somewhere in this resonator to smooth out the resonance peak or reduce the core saturation?
I also traced back the original article of that diagram and figured it out myself just now. Well, sort of.
So, I also saw the variation in the famous Opera M500 Modification that the tuning capacitor is in series with a 100R resistor. I suppose that is for tuning the disired Q of the resonator?
Then, again, can we use a diode somewhere in this resonator to smooth out the resonance peak or reduce the core saturation?
Note that the filter is only really effective at its resonant frequency (how much so and how narrow depending on Q) and wont attenuate noise other than the fundamental ripple very much.
Resonant filters.
Resonant filters.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.