|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.29.34.43
In Reply to: RE: an interesting journey posted by RayP on December 08, 2016 at 11:06:26
I have pure SMPS on the left channel on both the 813's and the 26, and I have halo on the 813's on the right but SMPS on the 26. I don't hear the edge on the left channel, only on the right, so my guess is it is the AC on the 813's that is doing it. Tough choice to make, and I will listen a while before I choose, but in the back of my mind is the other benefits of the pure SMPSno hum at all
no issues with frequency
no need to modify the equipment, just add filtering
sound that is a bit smootherI am leaning in that diretion.
Edits: 12/08/16Follow Ups:
Maybe it's a late entrance but I did follow what was said and done.
I would like to comment on the "edge", as well as on the "no hum" issue.
The "halo" as you call it (electronic transformers for halogen light bulbs) will not have ANY audible hum or artefacts - if the initial DC is properly filtered. The result is total silence when you stick your ear in the speaker. Yes, you might be able to hear some noise from your tubes, but that should be low, benigne and recognizable as tube hiss. Proper filtering depends on mains voltage and requires approximately twice the cap for US mains compared to European. Ray P already suggested 560u/200V but I missed the part where that was applied. Smoothing out this first DC is key and will lead to an increase in RMS output but will not change the peak voltage: the output transistors will not suffer but keep in mind the power equation and adopt oversized electronic transformers. 5A at 10V is 50W and I suggest to use a 150W rated unit for both channels (as shown on my blog).
Of course, with AC the hummer pot or better the pair of matched resistors is obligatory. It also cancels most of the 1st AC harmonic, so you are left with the attenuated 2nd. In this case assuming 30kHz, the 2nd is 60kHz: high enough for me.
As for the edge, there is none. I am used to the sound of AC on filaments of DH tubes. HF AC sounds the same, slightly cleaner (which I like and appreciate in any amp). I find DC to be muffled but to people who are used to DC because of hum issues with AC - DC is what they are used to. Let's try to skip the discussion about why is DC muffled ;)
One additional note is that my findings are related to my designs and amps based on those designs that I use and listen to. My RH300B is operated on 5VAC filaments and the hum is so low that one has to stick his ear in the woofer to hear a faint hum. The speakers are efficient enough at 93dB. On the TTA the hum can be heard a little better (but most would call it inaudible because it cannot be heard 1m from the speaker) with 2E22 which are DH 6.3V 1.5A pentodes (i.e. higher gain). I guess normal no-feedback SE amps are more prone to hum.
******
http://rh-amps.blogspot.com/
"Of course, with AC the hummer pot or better the pair of matched resistors is obligatory. It also cancels most of the 1st AC harmonic, so you are left with the attenuated 2nd. In this case assuming 30kHz, the 2nd is 60kHz: high enough for me."
The situation regarding harmonics is in fact even more benign than you suggest. If the wave is mostly square, very little even-order energy is present. The largest harmonic will be the third, or in the example you used, 90kHz.
The question for me is whether nulling the amplitude of the fundamental appearing at the anode is indeed sufficient to prevent unwanted mix products being generated within the tube. I haven't tested for this, thus I won't use a switching frequency below 60 kHz and would like to go much higher.
--------------------------
Buy Chinese. Bury freedom.
My whole point was that I got things adjusted to where the muffling really wasn't that bad, and implementing the SMPS unit was a lot less trouble. I still have not made a final decision, and since my 560uf caps just came in the mail yesterday afternoon, I still intend to try them and see how they sound. If the back ground noise and "edge" disappear I will certainly be more interested in staying with AC.
I am using one 5A 120 watt halo unit per 813.
Edits: 12/09/16 12/09/16 12/09/16 12/09/16 12/09/16 12/09/16
I can't find anywhere what you (vinnie2) wrote about the 5V and 10V filaments, but the following factors come into play when "taming" direct heated filaments:
1) voltage
2) current
3) tube gain.
Thus, a 2A3 at 2.5V (square: 6.75) is quite benign and easy to deal with. Besides, it has rather low gain.
A 300B at 5V (square: 25) is more difficult to deal with, but still relatively easy. The gain is also a little higher (if I'm not mistaking) than for the 2A3 but similar enough.
The 6B4G is basically a 2A3 on 6.3V filaments (square 39.69) thus it's more difficult to handle than the 2A3... but the gain is the same, i.e. low.
Now a 2E22 has 6.3V filaments too, but it draws more current (1.5A) and has much higher gain if used as pentode: as a result, a lot more hum than the 6B4G with the same filament voltage requirement.
The 813 has 10V filaments (square 100) and besides drawing 5A it has quite high gain! This literally means that it is very difficult to handle the hum on these.
I have tried operating the 813 on standard mains 10V AC, and the hum was very much audible in the whole room :( so this is definitely out of the question.
BTW the GM70 has 20V on the filaments (square 400 !) but it is not that much more difficult to handle than the 813 because the gain is much lower, being a triode.
You should not have issues with "edginess" but than again this could depend on the amp...
******
http://rh-amps.blogspot.com/
Hi Alex, I'm just curious; you make a point of stating the square of the heater voltage for each tube - what is the specific issue you are wanting to highlight there? (By the way, for the 2A3, the square of 2.5 is 6.25 not 6.75.)
Chris
Chris, it was a lapsus calami ;) 6.25 it is.
The noise is proportional to the square of the heaters voltage. That's one good explanation.
As for gain, again theory says the signal input at the cathode is amplified by mu-1 thus with pentodes that's way higher than with low mu triodes.
After all the hum is the mains frequency input at the cathode and amplified.
******
http://rh-amps.blogspot.com/
That's how this whole thing got started. I originally tried AC on the 813's and just could not tame the hum with conventional hum pots, etc. I was working with dc next, and a couple of folks mentioned SMPS and the halo units. One thing lead to another, and before you know it I was elbow deep in trying to get a quite 813 heater source. Believe me, I know what a bear the big transmitter tubes are for hum.
I stuck in the 560uf caps this morning, and although they have helped I still have some buzz and hum with the halo units. I will work on it some more and see if I can get it out of there. That is one of the best things about straight dc on the 813 and the 26, no hum at all. So why does dc have a muffling effect? : )
Edits: 12/09/16
In case you don't know, the lighting transformers have very poor line noise suppression. If your (AC) mains is clean, you are spared of the irony. If the mains noisy, the lighting transformer is not for you!
I 2nd for the option of the Rod Coleman Regulator with a really GOOD SMPS for filament use. Cheap and effective to get that near dynamic AC heated filaments.
But I still like Direct Heating with AC better. Direct heating with AC sounds more dynamic during the day where background noise is loud enough to mask that slight hum from the speakers. I switch to the SMPS with regulator at night.
.
.
.Thou shall not stand where I type for I carry a bottle of Certified Audiophile Air and a Pure Silver Whip.
I am trying to figure this stuff out myself, now when you say AC heated tubes sound more dynamic?! what makes them more dynamic??
Lawrence
That is why I am using filters after it leaves the SMPS.
Most good quality SMPS has good quality filter circuit built in and line noise isn't much of the problems.
It is the Lighting transformers that usually has the line noise problems.
It's hard to implement proper filtering after the SMPS. It's going to be high frequency so you need small value caps and what value depends on what frequency noise that you are trying to filter out.
My approach to SMPS is real simple. Get the best quality SMPS (trial and error) with very low ripple, and proper power sizing such as 4 x the wattage of your power requirement for the ripple to be low.
.
.
.Thou shall not stand where I type for I carry a bottle of Certified Audiophile Air and a Pure Silver Whip.
If you're right about this (mains impurity), that might be the most likely cause for the edge or glare.
I never had relevant issues with mains purity, although I don't think we have got particularly clean mains here in Belgrade. Thus I'm not inclined to accept that as a cause without further testing.
The buzz-noise-hum could easily be picked somewhere. If the other channel doesn't have it, try switching sides for the 2 heater supply types. If the buzz follows the culprit will be revealed.
If it's mains noise, would it be difficult to filter out some of it? Maybe a filter (cap and choke commercial type mains filter for appliances) would do the trick?
******
http://rh-amps.blogspot.com/
Are you expressing about mains noise in General? I am talking about the main noise rejection in the SMPS and Lighting transformers and in their specific applications for use in Tube filament supply.
Generally speaking, the main noise may or may not be the only problem in any audio system. AFAIK, an edginess is not the symptom of the mains noise but rather the whole amplifier / preamp and all other components.
It's hard to mention the effects of the mains noise in terms of the sonic in general. You can only compare when the mains is really clean and quiet and when it's noisy and the comparison of an overall effects of the noise in the AC line. This will lead us into never ending loop of the noise and the reciprocal effects in the audio system.
There are all kinds of induced noise in the SET system and the associated gears when used with highly efficient speakers. There is no easy one solution to address all. I for one just want to get my system as noise free as possible.
.
.
.Thou shall not stand where I type for I carry a bottle of Certified Audiophile Air and a Pure Silver Whip.
@megasat16 you said:
"Most good quality SMPS has good quality filter circuit built in and line noise isn't much of the problems.
It is the Lighting transformers that usually has the line noise problems."
I understood this as implying there's mains noise that gets filtered out with SMPS, but not filtered out with electronic transformers.
Therefore I have assumed you are accusing the mains noise for the hum-buzz etc issue.
Now I understand that you are actually stating your preference for SMPS over electronic transformers. I cannot agree with that based on the following:
1. I haven't experienced any hum buzz issues related to the application of electronic transformers. My results could be summarized as black and silent background. On the contrary, a car battery proved to be noisy, and all DC regulators I tried were noisy, too.
2. Even the SMPS option is bulky and heavy in comparison with the electronic transformer (modified by adding caps and changing the turns ratio on the output transformer). This comes as no surprise, just like the fact that electronic transformers are more energy efficient and disperse less heat.
3. Last but not least, SMPS is DC. This implies different operation with direct heated cathodes (difference in potential between the two ends), and all the subsequent sonic differences mentioned. I would like to skip the discussion about AC vs DC, as the "too detailed" vs "muffled" argument is a matter of taste and habit. Tastes and habits change but "de gustibus..." should be a nice civilized solution.
On the other hand, all electronic transformers for halogen light bulbs are not created equal and the end result is likely to depend on the model used.
Additionally, the math and skill might prove challenging to many, and measurements are not as easy as with any kind of DC. This doesn't stop me, but might deter others. My motivation would be rewarding sonic result and overall simplicity (and size) of this solution - not to mention the price.
******
http://rh-amps.blogspot.com/
I suggest you reread my previous post or all my posts on AA if time permits.
Your understanding of my preference of SMPS over lighting transformer is correct.
However, your assumptions of "I accuse buzz/hum is the reasons of main noise" is hugely incorrect and simply not what I said or meant under any context.
It could be a simple misunderstandings and misinterpretations on your own account.
I did say that SMPS have better line noise rejection / suppression than the lighting transformer.
However, I did not say or in any way imply that Line Noise causes the Buzz / or Hum / or Edginess anywhere in my previous posts. May be you mixed up my post with other posters' posts?
I don't know how or where you derive to that assumption but it is utterly and completely not what I meant. I simply give a comparison between a SMPS and the lighting transformers based on their ability to suppress / reject line noise.
Also, it is almost entirely UNTRUE that mains line noise will cause Buzz / or Hum / or Edginess in an amplifier or any other components in the audio system. If line noise could cause Buzz / HUM, it would be extremely (extremely) rare circumstances due to the extremely poor line conditions or some failed electronics (components) and house electrical wiring (switches) failure. I have never experienced such line noise problem that cause hum/buzz before.
Now, a mechanical noise from PT or choke is entirely a different matter from the power line noise! A power transformer may buzz/hum and that noise may be picked up by the nearby wiring under very rare circumstances but again it's not the same as Line Noise coming from the mains. A small DC present in the line could cause a Power Transformer to saturate under load that will emits noise. Also, an improperly mounted transformer may have mechanical hum/buzz but it's quite different from the noise in the AC mains! But all of these are not what the line noise coming through the mains.
There are all kinds of line noise that exhibits at various frequencies due to the induced noise (EMI/RFI) or from nearby equipment such as a running motor / compressor or a lighting fixtures or just some failed distribution /electrical components. Any of these noise could occur at various frequency and superimposed on the 50/60Hz AC signal thus causing a distortion in the AC sine wave.
How does that effect an overall sound system? There is no definite answer and the answer varies depending upon the equipment and the listener ears. That's why I don't want to get involved in the line noise problems in the first place.
Anyway, I consider lighting / led / electronic transformer are made for use in the lighting applications and not much else. The design is simple and parts are of so-so quality with no consideration required for line noise rejection or induced noise into the end appliances or any external interference (EMI, RFI). In it's defense, the lighting transformers are made to work with minimal parts count for low cost with OK efficiency.
On the SMPS front, there are countless designs and varieties for all kinds of applications out there. There are cheap SMPS with So-So quality parts and poor reliability but then, there are some really well made SMPS for use in medical and scientific use. And by design, even a modest SMPS has some kind of line noise suppression built in with load correction ability.
I don't have much to say about the quality of the lighting transformer or the use of such in an audio chain.
I don't know the effect of lighting transformers using for heating the DHT filament since I haven't really tried it.
It may work well but it may not. However, I don't care anything with the HF square wave (fundamental F1 plus countless harmonics) going through a series feed OPT in a SET amp which will end up in the GND and then into an endless loop.
Now, we shall discuss the remarks you made based on your experience.
1A. The lighting transformer may work well for you and many others. It may not work well for many (especially with those who owns very high bandwidth OPT). Also, many transformers will have a frequency response way beyond their quoted bandwidth even though the result could be highly undesirable and uncontrollable. I think you are finding an external interference with the battery analogy you have given. Some DC regulators can be noisy but audible switching noise means something wrong in the end design. All these noise should be fixable with an appropriate value decoupling cap.
2A. It's true that SMPS would be heavier and bulkier than the lighting transformer but it won't be much difference for the same kind of wattage. SMPS has advantages such as line noise suppression, blown out protections, load protection by self monitoring, thermal monitoring and self recovery under different load condition. Some SMPS can even reach 90%-94% efficiency.
3A. I agree that SMPS is not everyone. But it has potential when used with Rod Coleman Regulator. Having said that, I still prefer AC heating DHT filament by a small margin.
.
.
.Thou shall not stand where I type for I carry a bottle of Certified Audiophile Air and a Pure Silver Whip.
Only thing your missing now is a Rod Coleman supply. These are pretty sweet for the 26 tubes.
I suppose I will have to try one one of these days just to see what they are like.
I have a pair myself to try out between laptop dc supply 19v for the gm70 and the rod colman regulators with old school linear supply feeding them.
man these damn large final are true boat anchors!
Vinnie do you mind if i post up my gm70 BB build?
Lawrence
I misunderstood your question. Deaf has the right idea, a separate thread would make easier to follow.
Start another thread for your build. I have a GM70 build right now and can share my experiences.
Why would I mind?
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: