|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
61.18.170.94
If building a 300B SE, what would be more prefer for coupling between driver and output stage? Interstage transformer or capacitor coupled?
I've seen someone built an amp with 6922 DC coupled to 5687 and transformer coupled to 300B, but also have seen one with same tube complement but capacitor coupled. Pros and Cons between 2 different approach?
I heard transformer coupled is more efficient.
Follow Ups:
For builders, the Hammond 126C interstage is way better than the name "Hammond" implies. Bifilar wound, 106H inductance and sounds great. I abandoned capacitors a while ago except for using high inductance plate chokes with preamp tubes like the 26, where I use Russian teflons. And certainly plate resistors are a big no.
But biasing is an issue here. If you direct couple you usually have a large cathode resistor which then needs to be bypassed. I dislike cathode bypass caps - polypropylenes are worse than no cap and electrolytics suck all the life out of the music. I use the 126C with a 46 tube and no cathode bypass (2.1K resistor) - that's the cleanest solution I've found and the combination works.
andy
Direct Coupling: Dangerous cascading failure modes.. The upstream tubes grid bias condition is totally dependent on the downstream tubes plate voltage. That means if the downstream tube is unplugged, or its filament fails to ignite - you run the very likely risk of destroying the upstream tube. Sounds nice, no phase shifts like L or C coupling, but your opening yourself up to higher probabilities of catastrophic failure. Unfit for a commercial product design IMHO.
Capacitor Coupling: A leaking or failing cap can certainly induce failures but does not occur often. Phase shifts introduced by successive cap-coupled stages can setup the conditions for an unstable amplifier.
Tx coupling: Only failure mode is from a primary to secondary winding short. This would be caused by insulation breakdown. Very rarely happens.
An oddity if it does.
And you do know Wavac uses it and I do not see anyone complaining of failures with DC at all.
From what I've read in their literature and website, Wavac uses inverted interstage transformer coupling (IITC).
Jim,
In the USA market, here are three makers of single ended DC amps, two current, and one recent that I know of.
The FI amps, like the popular middle-priced "X" is a DC. Dennis Fraker's Serious Stereo amps have been DC since 1989. And the Moth amp's, well regarded, no longer made, are DC amps.
In all my experience in audio, I have NEVER even heard of a single failure ( due to direct coupling ) of any of these amps Jim. Have you sir ???
I have asked before about this, on this Forum, and no one has come forward and reported otherwise. I know as a fact, none of Dennis' Serious Stereo amps has ever "eaten tubes" due to a failure.
Jeff Medwin
.
.
.
My output tubes are biased around 60mA and I have 100mA fast blow fuses in series with them. I have no proof but I am confident the tubes can withstand the current for the time it takes the fuse to blow.
I currently have a 2 stage DC coupled amp and previously used a 3 stage one for many years with no problems. I know, this isolated case proves nothing but I have no qualms about dc coupling. I use components well below their rated values and protect things with fuses.
In any case the improvement in sound is well worth any additional risk if there actually is any.
.
,
The proof would be to do destructive testing of the amp, that would be a bit silly for a one off design.
I have blown the fuses on occasion and I have never lost a tube. Like I say, that limited data really isn't "proof" but if a tube can't handle 100mA for the length of time it takes a fuse to blow the tube has a problem anyway.
.
,
Regarding the Moth, Craig Uthus now sells a 2A3 dual headphone and stereo power amp, but I don't know if it's DC or not. Do you?
"You don't need to be a Weatherman to know which way the wind blows"
nt
"You don't need to be a Weatherman to know which way the wind blows"
People will most likely want to buy the original DC amps, used.
Jeff Medwin
nt
"You don't need to be a Weatherman to know which way the wind blows"
Isn't it ?? I was given the Moth schematic two years ago, and it is CLOSE to all the other DC amps I see. We easily can outperform these amps today !Jeff Medwin
Edits: 01/27/12
This is an application for over-current monitoring. If either tube goes into over-current for say one second, B+ will removed with a warning indicator LED of an over-current condition. I will install this on the DC amp I am building.
Hi!
There is no good straight answer to such a general question because teher are so many ways to do C-coupling and also so many ways to do transformer coupling. Chose the coupling method which fits best to the overall concept. Or optimize the concept for the coupling method you choose.
Best regards
Thomas
Both are easily bettered by direct coupling. Do a simple Loftin White two stage amp ala Shishido, Asano, with a good power supply..It makes sense, a piece of (good) wire will mess less with the signal than a cap or trannie to drive.
Two stage amps will always be more transparent sounding than three stages, KISS rules !!!
Cheers,
Jeff
Edits: 01/22/12
if you are DIY'ing, go for direct coupling. there is no way that any cap or IT can be better than 2 or 3 inches of wire.
Provide negative voltage for driver to zero your final's grid
Yes, direct coupled should be the way to go. Either stacked PS or like the DRD.
I heard different story, some says direct coupled need a stiff power supply with regulator. Even with 300B low powered, like the DRD, the B+ voltage will be the sum of driver + output. Now we are talking about 500+V. Regardless of big triode in DRD, almost impossible!! This is pretty tuffy to have a regulated power supply for DRD.
Another approach by talking to several tube builders and they prefer 50 or 70% more AC voltage on the secondary than the acutal B+ voltage and very low DCR secondary winding as well. Then drop the voltage from the first cap for actual needs feeding the low DCR & high inductance choke along with big bank of reservoir for proper filtering. This is the only way making a PS for a direct couple SE amplifier like the DRD. What do you think?
Hi,You state this :
"This is the only way making a PS for a direct couple SE amplifier like the DRD. What do you think? "
1What do I think?? Ha, you asked the right guy!
"This is the ABSOLUTE POOR WAY of way making a PS for a direct couple SE amplifier like the DRD. " No good.
Thats what I think .
I don't do DRD, don't want the cap in there where it is. I can do a Rk cap better than most, so its no problem for me. I prefer a simple a circuit as possible, Loftin White amps hold my interest the most. I prefer conservative tube dissipations and want ultra reliability to my builds.
I used to always regulate B+, no more. I like dual rectifier tubes and well-designed "modern" L1/C1/L2/C2 filters to the Finals.
A 2A3 DC amp needs 480 to 490 VDC of B+. A 6AQ5 DC amp (lower power) needs about 430 VDC. As for power supplies, email me privately with your own email address, and I will send you a couple of great papers as attachments, on what we are doing now a days, which is quite different from what you are thinking. Address above. Read them over, then ask questions !
Cheers,
Jeff Medwin
Edits: 01/25/12 01/25/12
.
At first I had problems with bias drifting along with the line voltage drifts. I am using stacked supplies so when B+ drifts with the line voltage the bias drifts.
Solution, use an AC regenerator. I bought a Crown pro audio 600W amp on sale, a low distortion sine wave generator, and a 3:1 power transformer hooked up backwards to the speaker outputs on the Crown to get a very stable, very clean, 120VAC. Added benefit you can control B+ and play with varying the input frequency.
I know there are off the shelf solutions but that's no fun and more expensive.
Oh, I can't generate enough juice to run the filaments too so they run off their own transformers hooked to the wall, not the regenrator.
.
Yes, direct coupled is a good idea. However, this is practical to some lowered SE amplifier, but not something like 845 or 211 that operate around 1100V for least distortion. Direct couple such as DRD on those big triode will require very high B+ voltage.
I would love to build a pair of 211 or 845 SET servo amps to drive a pair of ESL panels direct using no output transformer..
When problems become increasingly difficult,we can always rely on our collective knowledge and seek relief.
It's a good idea but better in PP. I have a trip in Europe and have seen Tim de Paravicini has done this back in the mid 90s for mod on Quad ESL panels (57 and 63).
Hey this is what I've seen people do, having the Audiostatic ES panel for 300 Hz and up and the stacked the Acoustat panels for the bass (crossover from 300HZ) and using the tube direct drive without the autoformer as you said. Sounds very powerful and clean, but you need a big room for those panels!!
BIG SOUND AND LOW DISTORTED BASS!!
I did something similar in a setup I had a couple of years ago. I was using Acoustat Spectra 2200 for everything above 100Hz (they made quite good bass already but not with that slam) and then I crossed that over to a pair of Acoustat Spectra 4400 for the bass < 100Hz. Those big Acoustat Spectra 4400 had AMAZING bass and slam (I did not direct drive though). The smaller Spectra 2200 sounded better with my tube amp (KR Audio VA350i) as they had an easier load. This made HUGE sound but with superb bass control and slam. Acoustic bass was simply the best.
Hi,
(1) If you need a 845 or 211 to power your speaker load, you will do better with a really good solid state audio amplifier, one that is of high quality.
(2) For someone asking such basic questions as cap versus transformer coupling, I would seriously caution you againt building amps with ultra high voltages for lowest distortion. ONE mistake with 1,000 to 1,100 VDC will likely kill you. It is very quick, permanent.
(3) A DC SE amp with 845s, low B+ is quite possible - 5 Watts output. Triode connect a 7788 and DC connect it to the grid of the 845. It will be more transparent, being a two stage, and will creme all the three stage amps out there, handily.
Jeff Medwin
Hi!
My friend has just showing me a fellow in other forum named Dowdy has buit a 2 stages GM70, and I saw his amp on youtube as well. And the other guy has intended to build also GM70 with Emission labs driver tube DRD coupled to GM70 with step up input transformer.
Maybe you are correct about the distortion issue from big triode, but doesn't mean 2 stages is impossible to work on big triode. Especially all Emission Labs's driver tube are DHT, and DHT driving another big DHT triode is pretty interesting if the quality of the step up input transformer can do their job well.
I read something about Dowdy's amplifier as this is a 2 stages with some big tube at front and parafeed. A pretty interesting approach as well. look, the most expensive Audio Note amplifier is a 211 driving by 10Y with "step up", and the LAMMM GM70 SE.
IMO, whether owning those amplifier is suicide is another issue. I used to build 845 back in the 90s with my best friend and never got it right since we have to parallel and series the lytic caps in the supply. We tried different circuit including the Audio Note Ongaku and Gakuon.
I'll not say big triode never sounding any good since I have listened to some fellow putting exotic double C amorphous output transformer made in Europe, I guess a company called AE along with high quality oil cap with 2KV rating, sounds pretty damn good. It sounds very different between those low powered and big triode.
Don't you know that running the big triode in low voltage making the driver stage more tuffy to drive. It requires much lower impedance.......so many things going to happen!! And the distortion retains on high side if operating at low voltage on 845 or 211. Why not just put a 300B for low voltage??
I also heard the people say the difference between graphite and copper palte from GM70. Well....they also sound very different. I don't build anything since the late 90s cuz getting older. But I highly respect those tube builder's "attempt" on big triode and the refinement they've done.
Nice talking to you!
Hello,Dowdy is a LONG TIME friend of mine, and I have had his amp, with Jim there, on my speakers in my living room.
I personally do NOT like a low or medium mu driver tube in a two stage amp, it makes it lifeless unless pushed hard to my ears.
You can triode connect a 7788 and pull off a big tube Finals amp OK, but if you really need that power, you have the wrong speakers, and /or should be using GOOD solid state, because it will outperform amateur builds I have seen. Overall, I prefer a 2A3 amp.
Jeff Medwin
Edits: 01/25/12
What you mean by low and medium mu? 50 or 100 or higher?
The high mu driver (in a two stage amp) is dependent partialy upon the mu of the finals.
To have a lively sounding amp at low listening levels, with a mu equals 4 or 5 (typical DHT) as Finals, I like to see about a 70 mu to 100 mu front end tube.
If you use a mu of 20 tube, (of which there are many nice sounding tubes ), it will never have the JUMP factor that I like to have in a DHT amp.
I am not into tonally correct but lethargic amps. I do not want to only hear it correctly when the amp is pushed. I want the amp, for long term listening pleasure, to be tonally correct and as lively and vivacious as all get out.
I need an email from you so I can have an address to send you the new power supply design data.
Jeff Medwin
The D3A I am using has a Mu of 70 in triode & a low plate resistance of 1900 ohms. Also seems to be more stabile than a 6C45n. The Pete Millet site have some interesting test data on the D3A. It has remarkable low distortion.
I am surprised I never seen a schematic with a D3A direct driving a 2A3.
Jeff, I like privacy- no face book, twitter, only Linkedin due to business ralationships. I was a ham radio operator just so I could experiment legally with a transmitted signal. I rarely ever called CQ, CQ, CQ. I communicated via aurora on 6 meters for the most part. I only socialize with women as I HAVE to to have one or attend business meetings.
Thats fine on privacy, I surmised that, thats why I suggested email(or snail mail) to a third party. How can I do that??
You, of all people, should read the grouping of posts.
Jeff Medwin
I have two addictions- The scent of a clean woman & the smell of hot tubes playing music. I like a glass of wine, but if I could only chose two, the above stands.
In my recent prototypes, I preferred capacitor coupling to an EI-cored, copper-wound interstage transformer. Capacitor coupling sounded higher resolution with more micro-details and micro-dynamics. The tables turn when you go to silver-wound, nickel cored ITs.
In my experience, silver is mandatory. I've never heard a copper transformer, be it MC step-up, TVC/AVC, interstage, or output, which had the clarity and resolution of a silver wound unit. Double C is probably also a requirement. My Silver Rock TVC uses Hi-B double C core transformers.
I'd love to listen to a silver interstage, but cost-wise, likely never going to be one sitting on anything I build. I'm sure they are a big step up. A couple years ago I bought a pair of the ITs used in the Audio Note IT Mono Blocks (regular EI core, not silver). So, over the years I have built many cap coupled, direct coupled and a couple IT coupled designs. I would say that direct coupled with a good PS is very hard to better, however you do face the issues of a lot of wasted heat and upper limits on the H+ (before getting into insane voltages). This means you're going to be looking at things like 2A3's if you want to keep to a 500v rail. Of the other two choices, I'd go IT any day and I'd say the AN EI core IT I have is better than cap coupled designs, I do not experience a lack of micro-details, etc... even though it isn't a top of the line IT. Cap is the last choice. Now a 4th choice here is "power drive" using a MOSFET follower. Check out TubeLab site for more on this.
In my Meishu, even the HiB (copper) dual c-core is a big step up from the standard OPT.
"You don't need to be a Weatherman to know which way the wind blows"
No argument, HiB double C core is an improvement over M4/M6 EI transformers. :)
nt
"You don't need to be a Weatherman to know which way the wind blows"
I drive ny 300Bs with Tango NC20F and the result is very very good! Better then a audio note silver coupling cap with grid choke, which is also very good. Anode choke with a top quality cap is also an option, but i think a good interstage like the NC20F is an investment for life. Good caps like silver foils also fail. Also know that cheaper transformers like the nc18 are not as good.
Regards Joćo
Edits: 01/22/12
I have built both. Although, I prefer IT coupling, RC coupling can sound great in a good design. The iron it takes to IT couple will run the BOM dollar amount way up.
How about cap coupling with grid choke loading?A good grid choke cost is a fraction of a good IT cost. There are some great sound PIO caps out there.
Stuben
Personally, I will never build another RC coupled amplifier. Power triodes need a low impedance path to ground in order to prevent "blocking distortion" where electrons build up on the grid upsetting the bias and causing distortion. IT coupling does provide the necessary path for grid current but introduces other limitations and challenges. A very good quality transformer is required for the best sound.
I believe (and everyone has their own opinion) that direct coupling solves the issues with both RC and IT coupling. For more information search on "DRD amplifier" and "300B monkey amplifier"
Below is the classic Komuro direct coupled amplifier with "stacked supplies."
.
I've built several stacked supply, direct coupled amps and couldn't agree more.
Direct coupled is the way to go.
.
Andy Gove has written on the topic of interstage drive extensively. The URL below was accessible, but you can google on Andy Gove + interstage transformers to find more complete statements by him.
My experience is that the sonic variation (good to bad)in cap coupled SETs is smaller than for transformer coupled SETs. The best transformer coupled amps I have heard sound better than the best cap coupled ones, but a number I have heard do not sound as good as cap coupled SETs.
This suggests that as easy as some transformer coupled amps look (see the circuit design of the Audio note kits interstage monos at http://www.audionotekits.com/monob01.htm), they are not as easy to implement as cap coupled circuits, and a lot depends on the transformer quality ($$$$$$$$).
"You don't need to be a Weatherman to know which way the wind blows"
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: