| Search | Login/Register
   Home » Audio Discussions » SET amplifiers (31 posts, 2 pages)
  Print Thread | 1st Post |  
Page 2 of 2 (31 items) Select Pages:  « 1 2
12-15-2014 Post does not mapped to Knowledge Tree
Paul S
San Diego, California, USA
Posts 2,656
Joined on 10-12-2006
Post #: 21
Post ID: 21407
Reply to: 21406
Don't Expect Too Much
fiogf49gjkf0d
I have been amazed that some people are not able to hear different operating points. And I have likewise been amazed when some folks liked the sound of their FCs with double the stock-design voltage. Of course I do not "recommend" that anyone vary FC voltage for any reason, including DSET matching. But anyone who can hear the differences might figure out for himself what is "appropriate" or "beneficial" in his own case, notwithstanding the idea that a perfect solution exists out of context.

Paul S
12-17-2014 Post does not mapped to Knowledge Tree
noviygera


Chicago, IL
Posts 177
Joined on 06-12-2009
Post #: 22
Post ID: 21408
Reply to: 21373
Loading the output stage
fiogf49gjkf0d
Hello Romy, This statement generated a question: >>With a "normal" set you have the output stage that let say loaded with 18x transformer.
What does this mean? Let's say the OT is 5K:8 where is this loading coefficient coming from? And how can it be adjusted? If the R(e) of a driver is let's say 12ohm, does this mean the secondary of the OT must be 12ohms for optimal loading to the given driver?
Gera

12-17-2014 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,155
Joined on 05-28-2004
Post #: 23
Post ID: 21409
Reply to: 21408
It means exactly what it means.
fiogf49gjkf0d
 noviygera wrote:
 What does this mean? Let's say the OT is 5K:8 where is this loading coefficient coming from? And how can it be adjusted? If the R(e) of a driver is let's say 12ohm, does this mean the secondary of the OT must be 12ohms for optimal loading to the given driver? 
 
I do not undesired what you ask. It means exactly what it means. If you have OT wish has a certain transformer ratio that you can change the transformer ratio and it will affect loading. A tube that you use has more or less constraint known output impedance. Well, it wary with the way how you power the tube plate but let presume that the amp is finalized and the operation point is fixed. The very small fluctuation of the output impedance with plate voltage is negligible at this point.  The impedance of you driver is knows, would it be 4, 8 16 R or whatever. Let say your tube is 150R and you decided to use 10X transformer. So, your driver of let say 16R will "see" tube of 150R x 10 = 1500R. If you change the impedance of your driver voice coil or ration of your transformer then the "loading" of your tube will be change as well. Do not forget that as much as driver "see" the tube via the prism of transformer in the same way the tube "see" the driver via the transformer.  What is the optimal loading of a given driver and a given tube? It is hard to say, you can measure distortion but it would not be too illustrating in some cases. Generally you use tube that everyone know and look what other people use. Thai is good place to start only and you will move from there. Some companies do easy re-mapable transformers section and you can do with them anything you want. 

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
12-17-2014 Post does not mapped to Knowledge Tree
JJ Triode
Posts 99
Joined on 09-12-2007
Post #: 24
Post ID: 21410
Reply to: 21409
Getting technical
fiogf49gjkf0d
Novygera:
If you have a transformer where the impedance primary-to-secondary is 5K:8, it means the turns ratio is (5000/8)½ = 625½ = 25.  This is the ratio by which the voltage is stepped down and the current stepped up in going from the tube to the speaker.  The output impedance is reduced by the square of the turns ratio.  Typically the optimum loading for power output (not necessarily Sound) is for the output tube to see a reflected impedance about 4x its plate resistance.  So 5K is optimal for a tube with a plate resistance of 1.25K.  As I recall, the plate resistance of a 2A3 is 800R and a 6C33C, 200R.

I don't know if Romy is talking about a transformer with a turns ratio of 18 or an impedance ratio of 18.  For sure though, the speaker driver should see the tube's plate resistance decreased, not increased, by the transformer.  For example, a lot of SET amplifiers have an output impedance of about 2R, corresponding to the "optimal" loading, and giving a damping factor of 4 into a nominal 8R driver.  For a turns ratio of 18, that translates to a primary impedance of about 2.5K which would be reasonable for, say, a 300B, so 18x is probably the turns ratio.

Rgds,  JJ
12-17-2014 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,155
Joined on 05-28-2004
Post #: 25
Post ID: 21411
Reply to: 21410
The irrelevant it is what it is.
fiogf49gjkf0d
I usually go with turns ratio. The ribbon tweeter has I think x175, the 6C33C full range is 10x or 12x and 6C33C in DSET is whatever it was – it is kind of irrelevant in DSET setting. The YO186/PX-4/RE604 I think has x25 transformer. My loading numbers shall be absolutely irrelevant to anybody as it reflects my drivers only. Also, my Midbass loading is very not optimum and was made in way "as is" – not the way how I would advise to do but in my case it was evolutionary process that juts ended up in the way how it ended up and I did not have already strength to change anything. 

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
03-24-2016 Post does not mapped to Knowledge Tree
noviygera


Chicago, IL
Posts 177
Joined on 06-12-2009
Post #: 26
Post ID: 22576
Reply to: 21411
Resuming
fiogf49gjkf0d
After planning out the exit strategy for my DSP crossover, I am resuming my first Single Ended Pentode 307a tube amp project, the one mentioned earlier. I'd like to approach this with a certain degree of understanding so I have re-read carefully all posts and comments and this time I feel I have less stupid questions to ask. As I understand, choosing the proper transformers for a dedicated SEP will give me ability to tune the amp-speaker relationship (and this is what I want). But I want to lay down, what I feel, may be constants in the context of this project so that I can wisely narrow down the variables.The schematic shows 5K primary impedance on OPT. My mid horn is used between 220hz and 1.9Khz and in this freq range it's impedance sweep curve is between 25 and 14 ohms.Does this info help me select a proper OPT? Is it safe to assume that secondary should have single tap 14-16ohm and primary may have, perhaps, three taps 4.5K, 5K, 5.5K that will leave me with the single variable necessary to tune the amp-speaker relationship? Am I missing additional variables that will be used to further adjust OPT-speaker relationship?Who makes re-mappable OPT?I found SILK transformers are reasonably priced and will make custom. But to order custom I need to have a good idea of how to specify "custom".
Gera
03-24-2016 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,155
Joined on 05-28-2004
Post #: 27
Post ID: 22577
Reply to: 22576
It takes time sometimes.
fiogf49gjkf0d
Gera, tuning the amp loading is a tricky part. If you deal tube that you do not know and the driver that you do not know then there is no way for you to know the entry point into the game. I think the best way in this situation is not use what people recommend out there or what transformer makers advise to use. Your schematics show 5K into 16R, so use it for now but make sure that it is against 16R. As you set the things up you begin to listen the thing and begin to feel where you need to go, make sure that at this stage your tube is locked in proper operation mode.. The increase load to the tube will give you richer, slower, more harmonic, less detailed and more anti-euphoric sound. To a degree you can test it by shunting the driver by a good resistor of 20-40R. Divide the response of your driver on 2-3 virtual sectors if you horn is between 220 and 2K then observe 220, 1000 and 2000 regions separately as they might be differently impacted by different loading. You might like lower knee with one loading but upper knee with another. If you less load tube then you get "faster" sound, with more transients, details and faster decay. There is no known to me ways to test it beside changing transformer ratio. Transformer taps might be good in here if they made properly (feeling the full bobbin) but I prefer the re-mapable sections. Lundahl makes very friendly transformers for such experiments...

A few warnings. As you change load make sure that you change the tube operation point and keep the tube in symmetric clipping. Idling tube more reduce power and reduce gain of your channel, so make sure that you re-adjust volume on the channel that you are trying to compare. Do not forget that your channel is not alone and it work accompanied with other channels. In some instances, an particularly in your where there is no HJF in band bath, it is preferable to have a channel to be driven a bit too hard (more idle tube)  and the shortage of harmonics at lower end to compensate by some overlapping of the next channel at the bottom. 
 
It feels as your channels all have to be load in the same way for own reference impedance but it not might be the case. You need to load each channel in order to get a complete presentation from all channels that you approve. That might take time. Sometime only after a year or two you “get” that one channel might be driven more aggressively…
 

Rgs, Romy

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
03-24-2016 Post does not mapped to Knowledge Tree
JJ Triode
Posts 99
Joined on 09-12-2007
Post #: 28
Post ID: 22578
Reply to: 21406
SEP loading and feedback
fiogf49gjkf0d
Novygera,
If you are really going to run a single-ended pentode as a pentode (not triode-strapping the screen grid) then you will almost certainly need to use some negative feedback, whether global or local.  It is partly needed to reduce distortion, at least that is "common wisdom" but even more needed to reduce the output impedance, which is very high for a pentode running as such; like roughly 10-20 times higher than a triode, or triode-strapped tube.  Feedback gives you another variable to play with, and the subjective effects are somewhat similar to what you get with what Romy calls reduced or lighter loading: more feedback = more transients, "tighter" sound but eventually more hardness.  At least this is true for voltage feedback, such as globally from the output positive tap to the input tube cathode.  There is also such a thing as current feedback, which may sound differently.

SEPs are kind of strange animals.  There is a lot less information about them out there than for SETs.  You can find some discussion on local (like the RH amplifiers) versus global feedback, and screen power supplies (gas tube regulation at significantly lower voltage than the plate is advocated by some.)
Good Luck,
JJ
03-25-2016 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,155
Joined on 05-28-2004
Post #: 29
Post ID: 22579
Reply to: 22578
No pentode intended
fiogf49gjkf0d
I would like to point out that whatever I said might or might not be applicable to amp with pentode in output stage. I have absolutely no experience with driving speakers with pentode and I never used any feedback in my tube amps. So, I just do not know if anything that I know (neither I care) has any relevancy to pentode amps.

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
03-26-2016 Post does not mapped to Knowledge Tree
noviygera


Chicago, IL
Posts 177
Joined on 06-12-2009
Post #: 30
Post ID: 22580
Reply to: 22579
Mappable transformer
fiogf49gjkf0d
Romy,
I think I'll try your suggestion. When you say "mappable transformer" do you mean one with multiple primary taps, like 3.5K, 5.5K, 8K?Gera
03-27-2016 Post does not mapped to Knowledge Tree
Romy the Cat


Boston, MA
Posts 10,155
Joined on 05-28-2004
Post #: 31
Post ID: 22581
Reply to: 22580
Many of them.
fiogf49gjkf0d
Many transformers have multiple sections for primaries and multiple sections for secondary. You can map them in any way you want, parallel or in series, or even if you want you can love a section unused which is controversial and I am not widely support. Any transformer maker know most of the tube and will give you a good starting point to experiment. For instance Lundahl is represented in US by K&K audio and Kevin Carter who runs the shop will be very instrumental for you to figure out what you need.

"I wish I could score everything for horns." - Richard Wagner. "Our writing equipment takes part in the forming of our thoughts." - Friedrich Nietzsche
Page 2 of 2 (31 items) Select Pages:  « 1 2
Home Page  |  Last 24Hours  | Search  |  SiteMap  | Questions or Problems | Copyright Note
The content of all messages within the Forums Copyright © by authors of the posts