A large capacitance in the 1st position of a CLC filter is out because the startup surge causes arc over and device destruction.
Use the minimum capacitance in the 1st position consistent with keeping the rail voltage up. Lots of capacitance in the 2nd (reservoir) position is fine, as the choke isolates both the rectifier and the input cap. from the 2nd cap. and any additional elements present. BTW, this is good procedure with SS diodes too. Just because a SS rectifier can tolerate a gargantuan valued cap. I/P filter, doesn't mean such a cap. should be used.Large capacitance is needed in the stack of a Greinacher doubler. A really black background in this situation requires a RF "hash" filter between the doubler stack and the remainder of the PSU filter. Apply Fourier's theorem to the highly "triangular" ripple wave form.
A large capacitance in the 1st position of a CLC filter is out because the startup surge causes arc over and device destruction.
Eli,
That was what I meant but didn't say, downstream of the rectifier tube caps can be as large as you would want. It is possible that some people more familiar with SS do not fully understand tube circuits and data sheets. The "cap input" value is just the max or recommended cap value immediately following the rectifier tube.
As to the SS vs. Tube rectification issue that the OP asked about. His intention is to build or buy a 300B amp. He is simply asking about SS vs. Tube rectification.
IMHO, I suggest to build the amplifier, there is no more satisfying experience than to do it yourself.
If built as 2 mono blocks your looking at a minimal current requirement easily supplied by a 5U4GB or 5R4GY and most power trannies will have have plenty of 5V current to supply the rectifier and the 400B filament.
Schottky diodes are an option but again IMHO, do not belong in a "classic" SET amp of this design.
"Soft Bass" is again just a product of the Power Supply design. Search for 300B SET schematics that have well filtered highly capacitive power supplys and you should be fine.
Here is a CLC filtered version
6SN7 / 300B Single-Ended (SE) Tube Amplifier Schematic
I would suggest playing with the PSU design slightly for larger C value in second postion.
Also the DC filament supply for the 300B could be better filtered
and another with AC filaments and similar PS
300B Mk1 Tube Amplifier
and look here for a pleathora of 300B designs
MachMat tube schematics
6SN7 / 300B Single-Ended (SE) Tube Amplifier Schematic
I would suggest playing with the PSU design slightly for larger C value in second postion.
Also the DC filament supply for the 300B could be better filtered
and another with AC filaments and similar PS
300B Mk1 Tube Amplifier
and look here for a pleathora of 300B designs
MachMat tube schematics
Hey All,
Happy New Year, for a start. I am building my first amps with tube rectification. I was not familiar with the fact that tube rectifiers have a limit on the supply caps until recently. The GZ34 has a limit of 60 uf, and the 5U4GB 40 uf. Having looked at a few schematics I've found the first cap to be the largest. And the remaining caps even, say 40,20,20,20. That doesn't seem to jive with Eli's above post? Putting a smaller cap in the first position. Is that true because of the CLC?
And when the spec says C 60 uf, that means the largest cap is limited to 60 uf, not a combined total? Just wanted to be sure.
And, as an aside, can HV supplies be regulated? I think I remember reading somewhere that they shouldn't be? Couldn't be? But I can't seem to find that post. I would like to use the GZ34 because it allows greater capacity/filtration, but I need the 50 volt drop of the 5U4GB to use the transformers on hand.
Kevin
Happy New Year, for a start. I am building my first amps with tube rectification. I was not familiar with the fact that tube rectifiers have a limit on the supply caps until recently. The GZ34 has a limit of 60 uf, and the 5U4GB 40 uf. Having looked at a few schematics I've found the first cap to be the largest. And the remaining caps even, say 40,20,20,20. That doesn't seem to jive with Eli's above post? Putting a smaller cap in the first position. Is that true because of the CLC?
And when the spec says C 60 uf, that means the largest cap is limited to 60 uf, not a combined total? Just wanted to be sure.
And, as an aside, can HV supplies be regulated? I think I remember reading somewhere that they shouldn't be? Couldn't be? But I can't seem to find that post. I would like to use the GZ34 because it allows greater capacity/filtration, but I need the 50 volt drop of the 5U4GB to use the transformers on hand.
Kevin
Hi Kevin,
Best wishes too.
As you are building amps yourself, would not it be good to stretch your knowledge on basic things a bit? This is what comes up reading your questions.
DIY constructing is great, but often it seems that a bit more DIY on basic knowledge is lacking.
There is much much to learn from forums, and there are very good books on most of the basic subjects. Read Morgan Jones for instance.
Best wishes too.
As you are building amps yourself, would not it be good to stretch your knowledge on basic things a bit? This is what comes up reading your questions.
DIY constructing is great, but often it seems that a bit more DIY on basic knowledge is lacking.
There is much much to learn from forums, and there are very good books on most of the basic subjects. Read Morgan Jones for instance.
Rest assured you're not alone. "Respected" manufacturers have used 300uF even 1000uF immediately after the rectifying tube.
One thing you simply must do is start using PSU Designer II. It takes 10 minutes to install and figure it all out and it will answer all your questions and more.
If a choke is large enough it will effectively isolate the capacitance from the tube by limiting the peak current. So will obviously a resistor but at the expense of dissipated power, lost voltage and raised dc component of the supply impedance. Thirty years ago, not having access to chokes i built the best power supply i could imagine for my phono stage. Rectifier was an EZ81 and after a couple of RC smoothing stages i had a gargantuan 3300uF/350v followed by a final RC stage with much smaller capacitance. It had practically no ripple or any noise even without chokes, took nearly 10 minutes to fully charge the cap and sounded amazing. Should try it out again sometimes and see how it compares to shunt regs.
Thanks,
I have some books that Eli suggested on the way but they haven't arrived yet. That program looks really useful, thanks. As with any technical endeavor its the little things you don't know that bite you. And though I have read many threads here on DIY sometimes its hard to remember where the information was first seen. And, in my own defense, it isn't like there is that much consensus here. But then that's what makes it interesting!
Kevin
I have some books that Eli suggested on the way but they haven't arrived yet. That program looks really useful, thanks. As with any technical endeavor its the little things you don't know that bite you. And though I have read many threads here on DIY sometimes its hard to remember where the information was first seen. And, in my own defense, it isn't like there is that much consensus here. But then that's what makes it interesting!
Kevin
"Respected" in the high end does not equate to competence. Using a cap that large is a near guarantee of unreliability with no performance gain. Which, I guess, is part of something being considered "high end"- if it functions flawlessly and reliably, it's suspect.
Kevin,
I too was very eager to "jump in feet first" on my first build.
A few "wise sage's" here on the forum "set me straight" on how to approach this.
#1. At the VERY LEAST download an RCA tube manual. I suggest RC30. Available for download all over the web. (Pete Millett's site for instance) Read the sections at the beginning and understand what is going on.
#2. Morgan Jones is like a "guru" read those books and papers.
#3. Download PSU Designer!!!! I use it as a bible, especially when using "on hand" or "working pulls" for transformers.
#4. Do not worry about the voltage drop of the rectifier tube, more importantly that it can deliver the needed current. You can easily add another RC pole to drop the voltage and add capacitance and more filtering. (You can play with PSUD to see what I mean)
#5. A quick "rule of thumb" for the input cap, go relatively small maybe 1/2 of the max in the data sheet to start with. If this gives you an acceptable B+ work from there. Need more voltage go bigger, too high? go smaller.
#6. There are some calcs on the web that will show you the amount of 120Hz noise attenuation that a combination of R & C will give, use those to calc the RC poles from there. Later, as you learn more you can get into LC filters.
#7. You CAN regulate B+ but I would only look into regulating the preamp stages of the amp. (VT tubes work for this but there are noise issues requiring attention same with Diodes.)
I too was very eager to "jump in feet first" on my first build.
A few "wise sage's" here on the forum "set me straight" on how to approach this.
#1. At the VERY LEAST download an RCA tube manual. I suggest RC30. Available for download all over the web. (Pete Millett's site for instance) Read the sections at the beginning and understand what is going on.
#2. Morgan Jones is like a "guru" read those books and papers.
#3. Download PSU Designer!!!! I use it as a bible, especially when using "on hand" or "working pulls" for transformers.
#4. Do not worry about the voltage drop of the rectifier tube, more importantly that it can deliver the needed current. You can easily add another RC pole to drop the voltage and add capacitance and more filtering. (You can play with PSUD to see what I mean)
#5. A quick "rule of thumb" for the input cap, go relatively small maybe 1/2 of the max in the data sheet to start with. If this gives you an acceptable B+ work from there. Need more voltage go bigger, too high? go smaller.
#6. There are some calcs on the web that will show you the amount of 120Hz noise attenuation that a combination of R & C will give, use those to calc the RC poles from there. Later, as you learn more you can get into LC filters.
#7. You CAN regulate B+ but I would only look into regulating the preamp stages of the amp. (VT tubes work for this but there are noise issues requiring attention same with Diodes.)
The lack of consensus on basic facts is precisely the reason for seeking your own learning. There is even a certain consensus in audio circles around some ideas that are provably incorrect (using science).
People like to sound knowledgable and will often repeat things they have heard. The internet is very effective at promoting this behavior. Often this seems like a genuine attempt to be helpful, but ultimately if people would just stick to their own deep knowledge there would be a lot less confusion.
Many manufacturers of "modern" tube gear didn't grow up with tubes and have acquired the knowledge second hand. I think this is where ridiculous designs like cap input 470 uF filters on tube rectifiers come from. All you need to do is read the %&$^ datasheet!
A lot of DIY builders don't care about the real science behind it either and I suspect many "want to believe" in the magic. It can be a lot of fun and what's the harm as long as no one gets electrocuted
So I contend that this forum and others like it are not effective venues for real learning to take place. Generally one needs to go back to material written when facts mattered in vacuum tube design (through about 1970) and learn that material first, thus forming a factual basis from which to evaluate the "new knowledge".
Happy new year!
Michael
Most of all ENJOY the process!
Other things to look into:
LTSpice
Tube Cad
Glass Ware's website
I recommend looking at building a commonly built simple SET or PP amp. Try to source parts as close to the original schematic as possible but you can deviate as you learn.
For best "satisfaction" keep it simple, a williamson style PP amp or an IDHT SET would work nice. I am not sure of where you are headed but a quick post with your ideas, expected results and what "on hand" parts you have will certainly garner you some informative replies, SY and ELI particularly can offer help.
Use google and look for amp schematics using your choice of output tubes.
IE: "6V6 Push pull amp schematic"
Other things to look into:
LTSpice
Tube Cad
Glass Ware's website
I recommend looking at building a commonly built simple SET or PP amp. Try to source parts as close to the original schematic as possible but you can deviate as you learn.
For best "satisfaction" keep it simple, a williamson style PP amp or an IDHT SET would work nice. I am not sure of where you are headed but a quick post with your ideas, expected results and what "on hand" parts you have will certainly garner you some informative replies, SY and ELI particularly can offer help.
Use google and look for amp schematics using your choice of output tubes.
IE: "6V6 Push pull amp schematic"
DITTO!!!
EXACTLY why I suggest the "older info" like Morgan Jones and RCA Tube manuals. Also, look at schematics of long obsolete items to see what was done while tubes were the "state of the art".
I am only 40 years old and most of the stuff I have learned from in this hobby has been obsolete before I was born.
What Michael says is true, just because XYZ company uses HUGE input caps does not mean it is correct, these "designers" are simply regurgating something that pretty much has been done blindly.
"Back in the day", large companies were spending huge amounts of money time and effort to research and test what they designed. That is where the DATA SHEETS come from, (Not theory or conjecture) but real world testing. USE THE DARN DATA SHEETS!
Yes, they have. They were scamming a largely atechnological public which doesn't understand that having to replace the 5U4s or 5Y3s every six months or so is abnormal. It also didn't make much difference when every pharmacy, green grocer, five-and-dime, etc, all had those "test 'em yourself" set-ups, and you could easily get another 5Y3 for a couple of bucks or less.
These days, they're not so easy to come by, and we'd like to get some usage out of them. You also don't want to over stress power xfmrs that weren't designed with such excessive Isurge in mind either.
Build as cheaply as possible, sell low, and make up the difference on volume was as operative back then as it is today.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.