DIY Audio Projects Forum
 NEW  Matt presents bias and operation data for the 6V6 tube in SE operation - 6V6 Single-Ended (SE) Ultra Linear (UL) Bias Optimization.

DIY Audio Projects Forum

Welcome to the DIY Audio Projects Message Forum. Use these forums to discuss Hi-Fi audio and to share your DIY Audio Projects. Registration is free and required to post messages and view the file attachments. Registration will only take a minute and registered users do not see any advertisements. After you have completed the online registration process, check your email (including spam/junk folder) for the verification email to activate your account. New members are under moderation - so your posts will not be visible until approved by a moderator. See the Read Me 1st, Forum RULES and Forum FAQ to get started on the forum.

It is currently 16 Oct 2019, 01:21

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 84 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8, 9  Next
Author Message
PostPosted: 22 Aug 2010, 01:00 
Offline
Project Author
User avatar

Joined: 06 Apr 2009, 10:08
Posts: 1592
Location: US Pacific Northwest
I am very interested in seeing how the SE ATS25/807 Two Stage amp (http://diyaudioprojects.com/Forum/viewtopic.php?f=9&t=2265) turns out. It has also got me to thinking about SE designs in general.

I have been thinking that the one thing that might prevent people from building this particular design is the high bias current level and consequently the large and expensive power and output transformers. I have been rereading the STC807 report and thinking about how to lower the bias current draw using this tube but still preserve the characteristic smooth and expressive SE sound. This would allow for less expensive power and output transformers and hence a cheaper overall build. I have also been thinking about something mwhouston said about the SE amps seeming to sound better when the tube is driven hard. (i.e. close to its plate dissipation limits) This has been my experience as well.

All of this got me to thinking about how we specify SE performance and what figures of merit we use. Looking at various beam power tube data sheets that show triode strapped operation, they all seem to have a common characteristic. They tend to keep the THD numbers fairly small and they all seem to run the tube far below its plate dissipation limits yielding fairly low peak power output numbers. This is understandable for two reasons. First no designer wants to specify a large peak distortion number; this makes the tube look bad. And second, if the object is to sell beam power tubes and not triodes, showing very low triode output power verses the tube run in pentode mode, shows the tube in a better light. However, since I’m not interested in selling tubes but rather in getting a particular sound out the tube, I probably want to do things a little differently.

Now in SE amps the distortion tends to be approximately proportional to the output power. And since the peak output power level is different for each design, the peak distortion numbers published in data sheets really don’t have any value for comparing different designs. So I have decided to begin looking at a new figure of merit for SE amps and this is % THD per watt. I have started labeling this “Ds” in my worksheets. In this way, if I have one design that is 4%THD at 2.2W (Ds=1.8%/W) and another that is 8%THD at 4.5W (Ds=1.8%/W) I know that they will have similar distortion characteristics at similar listening levels (all else being equal).

Getting back to the 807, I took a look at the “typical” triode mode data in the STC807 report. The fixed bias condition shows a bias point of 250v at 40mA (10W plate dissipation) yielding 1.3W peak output power at 4.5%THD (Ds = 3.5%/W) into a 5000Ω load. For cathode bias at the same bias point, is shows 1.1W peak output power at 4.0%THD (Ds = 3.6%/W) in to a 6000Ω load. This seems to me to be a fairly anemic way to run a tube rated for 25W plate dissipation. So I decided to try a few other options. (See the attached file.)

I started by choosing two different bias points using the same 5kΩ plate load suggested in the report and at similar (but higher) plate dissipation levels. The first bias point is 390v at 42mA (~16.4W plate dissipation). This yielded a peak power output of 4.1W at 9.2%THD. This 9.2% distortion may sound bad, but Ds is only 2.24%/W. This means that this design will have approximately 36% less distortion then the one suggested in the report when played at the same power level. Additionally I have gained tremendous headroom in this design in comparison to the “typical” design suggested in the report. The second bias point is on the same load line but is at 347v at 51mA (~17.7W plate dissipation). This yielded a peak output power of 3.0W at 7.1%THD. Here Ds is 2.37%/W. So the similar load line and bias points may have yielded different peak powers and distortions but the distortion per watt “Ds” values are very similar.

I then decided to try the same 5kΩ load line but push the tube much closer to its limits. This bias point is 370v at 68mA (~25.2W plate dissipation). Now I know that the 807 data sheets say the peak plate rating is 25W, but this is for Continuous Commercial Service (CCS). The assumptions behind CCS are 24/7 operation and a continuous modulated RF carrier power of 25W. In Intermittent Commercial and Amateur Service (ICAS) the 807 is rated for 30W plate dissipation. This assumes intermittent operation and non-continuous modulation power so I don’t feel bad at all designing an audio output stage with 25.2W peak plate power dissipation. This is well within the 807’s ratings envelope.

With this load line I discovered something very interesting. This load line yielded a peak power of 3.3W (right between the previous two designs) but at a 4.8%THD. This yields a Ds value of only 1.45%/W! This design will have approximately 58% lower distortion then the “typical” design given in the report and has almost two and a half times the headroom. In other words, this design should sound much better then the suggested design. It is also very interesting that the design which drives the tube hardest is the one with the lowest Ds value.

So what does all of this mean? I think there are several conclusions we can draw. First, the 807 really does like to be driven hard. If it’s triode characteristics are typical of triode strapped tetrodes and beam power tubes (and I think they are) then we can extrapolate this observation to other tubes as well. Second, the “typical” triode strapped designs given in data sheets must be evaluated carefully. In all probability they are not optimal for the tube in question but are chosen with “other” characteristics in mind. And third, the comparison of different designs requires a figure of merit that takes in to consideration the characteristics of the SE design (i.e. output power vs. distortion proportionality). I believe that the “Ds” parameter helps to do this.

Comments anyone?


This post has a file attachment. Please login or register to access it. Only Registered Members may view attached files.


_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog


Top
 Profile  
 
PostPosted: 22 Aug 2010, 05:39 
Offline
User avatar

Joined: 18 Feb 2010, 05:14
Posts: 48
Location: Incheon, South Korea
Boy am I glad to hear that. The SE KT88 amp I'm building is right up against its max plate dissipation curve.

From reading about George's 845SE amp and his powerdrive circuit, he seems to agree that running tubes close to their maximum plate dissipation values is the way to go.

I played around on SE Amp CAD with the circuit I'm using, and it was consistent with your findings. "Ds" decreased the closer the simulation was to the maximum plate dissipation curve.

_________________
A Paragon of Mediocrity Since 1982


Top
 Profile  
 
PostPosted: 22 Aug 2010, 06:44 
Offline
Project Author
User avatar

Joined: 06 Jun 2008, 18:23
Posts: 5139
Location: Australia
Guys I am glad to read your findings and remarks. I started pushing SE KT88s and felt I got more from them without a single data sheet or distortion analyser in sight. My ears simply said "that sounds better". The big revelation came when I followed Bruce into the KT120 camp. The harder I pushed these to beyond the limits of my power tranni and OPTs the better they sounded.

Then came the 6AS7 SET amp. By this time I was building amps with a soldering iron in one hand and whip in the other. I had made three amps and lots of mods in just a few month's period so I was hearing and comparing the differences at every turn.

By accident the the HT ended-up 20V over on the 6AS7 amp and drove the old English United Electronics tube harder than intended but it sounded fabulous. I swaped the pommy tube for a new Winged "C" and due to different tube characteristics the poor thing was a "burning amp" waiting to happen. The tube was dissipating 14W!!. I was convinced to throttle it back a bit. Even I couldn't be that cruel. I dropped the HT+ 20V which lowered tube diss. to just under 12W (tube rated 13W).

I lost no audible performance by just being a little kinder on the tube. BUT Suncalc made a very importing point when he said the rated diss. for a tube is stated with the belief that the tube will be on 24/7. Now I play my main system a lot but on average 4/7 would be about my weekly system up-time. The tube may out live me even driving them harder than most tubaholics.

Bruce is much kinder on his tubes looking for about 80% max diss. But as I have stated Bruce produces an excellent commercial product and purchasers if his gear have a right to expect good tube life. If I burnt a couple of tubes a year for what I believe is a better sound from my SETs then so be it. At this point I will continue in my dark ways.

After saying all of that I did read in the latest Stereophile that gently pushed, quietly idling SETs were considered to produce the best sound. Maybe they have never heard a SET in a deep sweat!!!!

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp


Top
 Profile  
 
PostPosted: 22 Aug 2010, 10:48 
Offline
Project Author
User avatar

Joined: 06 Apr 2009, 10:08
Posts: 1592
Location: US Pacific Northwest
Well, I've bought a couple of 807s and I think I'm going to try prototyping a few different output stages to see how they sound. I'll order a couple more 5kΩ Edcor GSXEs this week. (I can also use these in a 6CY7 or 6EM7 spud amp that I'm thinking about as well. Single ended of course.) I want to verify experimentally what the math seems to be telling me.

I'll probably get around to this as soon as I'm done with my current 6V6 UL SE stereo amp which I'm currently putting together. (Albeit slowly. I'm spending more time with a calculator in my hand than a soldiering iron these days.) I'll keep everyone posted on my progress.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog


Top
 Profile  
 
PostPosted: 03 Sep 2010, 16:41 
Offline
Project Author
User avatar

Joined: 06 Apr 2009, 10:08
Posts: 1592
Location: US Pacific Northwest
Well, the Edcor transformers I needed to tryout my "Ds" parameter theory are at least 5 weeks out according to the manufacturer. Looks like I'll have to be content getting my latest 6V6 SE UL amp put together.

If anyone has a need for Edcor transformers comming up, you might want to order now. They currently have an order backlog that they are trying to fill.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog


Top
 Profile  
 
PostPosted: 03 Sep 2010, 17:24 
Offline
Project Author
User avatar

Joined: 06 Jun 2008, 18:23
Posts: 5139
Location: Australia
Suncalc wrote:
Well, the Edcor transformers I needed to tryout my "Ds" parameter theory are at least 5 weeks out according to the manufacturer. Looks like I'll have to be content getting my latest 6V6 SE UL amp put together.

If anyone has a need for Edcor transformers comming up, you might want to order now. They currently have an order backlog that they are trying to fill.

Edcor, I think, only do a run of a certain tranni every now and then. They don't appear to keep stock. They may also only make to order and wait until there is sufficiant orders to do a run. Frustrating!

I have now been waiting 6 weeks for some power and OPTs. I only just recieved the notice that they have shipped. Also their administration area appears disorganised. After waiting weeks for news of my order shipping they claimed I hadn't paid through PayPal. I never got the PayPal notices. Even after checking my PayPal account there was no mention of the request for money. So I PayPaled them!! Once I was ambused by Edcor for replying to an e-mail but it was the wrong e-mail address for ordering gear. Thought they e-mailed me first about my order??!!??

Now on the upside. The tranni's are good and well priced. So SUCK IT UP!!

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp


Top
 Profile  
 
PostPosted: 06 Sep 2010, 05:43 
Offline

Joined: 07 Feb 2010, 15:42
Posts: 151
Location: England
mwhouston wrote:
The big revelation came when I followed Bruce into the KT120 camp. The harder I pushed these to beyond the limits of my power tranni and OPTs the better they sounded.

Bruce is much kinder on his tubes looking for about 80% max diss


My question, albeit more offtopic than ontopic, but related to your post, is this...

On Bruces oddblock build with the ccs, If I wanted to drive the tubes into the ground (so to speak) I would just have to push more mA into them by dropping the resistor value?

Or does the overdriving of tubes really only work on an SE amp?

Thanks :D

Lewis

_________________
:D


Top
 Profile  
 
PostPosted: 06 Sep 2010, 07:32 
Offline
Project Author
User avatar

Joined: 06 Jun 2008, 18:23
Posts: 5139
Location: Australia
Because I do not use CSS on my SETs Bruce has all the answers.But I can't see why CSS based amps/tubes cannot be pushed just like non-CSS types.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp


Top
 Profile  
 
PostPosted: 06 Sep 2010, 14:24 
Offline
Project Author
User avatar

Joined: 06 Apr 2009, 10:08
Posts: 1592
Location: US Pacific Northwest
I have not had a chance to verify my suspicions experimentally, so I am talking only from theory here.

The occurrence of reduced Ds in the SE amps as the bias point approaches the operating limit curves is based on the typical curve characteristics (triode and UL, primarily linearity) as load line moves away from cutoff. In an SE amp, the AC load line passes through the DC operating point thus the AC operation stays in the vicinity of the operating point. (i.e. the reactive ac ellipse is centered on the operating point with the major axis aligned with the AC load line.) It also means that we can be assured that the operation stays away from the cutoff region.

In a PP amp you have a different situation. Here the AC load line does not pass through the DC operating point but through zero current at the DC operating voltage. This much steeper load line runs directly through the cutoff region. As such, the only way for the reactive AC ellipse to stay up near the tube limits is to limit the operating class of the amplifier. This means increasing the bias point and keeping the drive voltage such that both tubes only operate in pure class A (i.e. both tubes conducting current through the entire cycle).

Of course, the fly in the ointment so to speak, is that the distortion in the SE amp is approximately proportional to output level. In a PP amp, the distortion is relatively fixed by the operating point placement and the steepness of the load line. Thus where the low Ds value in the SE amp make's it sound very clean at low volumes and then only shows additional distortion as it is pushed to higher output, the PP amp show basically the same distortion at all volume levels with the distortion going from mostly odd harmonics at low volumes to even and odd harmonics at higher volumes.

The result of all this is that I think you should get the similar sonic improvements in PP amps by pushing the tubes harder, but I believe that you'll only see the improvements if both tube stay in class A all the time. I think that once the tubes begin to approach cutoff and the amp transitions to AB, any linearity improvements you've obtained by pushing the DC bias point will be lost.

_________________
Matt
It's all about the Glass!
http://www.CascadeTubes.com
Cascade Tubes Blog


Top
 Profile  
 
PostPosted: 06 Sep 2010, 16:58 
Offline
Project Author
User avatar

Joined: 06 Jun 2008, 18:23
Posts: 5139
Location: Australia
I just read an where the Audio Space monoblock 300B amps have input bi-wiring provision. It's a new one on me. The way I see it is you can run two sets of interconnects from the preamp to the power amp. Not dual inputs for two different sources but bi-wire the one source?

Put your hand up if you have heard of this before? I'll post results later.

Suncalc: One thing I have gathered from your post above is PP amps may sound better at higher volumes where the harmonics are predominately even where SETs sound better at lower volumes where distortion is at its lowest. I think the lessen here may be to ensure both amp topologies stay in Class A.

As an aside: Even after weeks and weeks of listening to my 6AS7 SET I am still marvelled by how good it is from a detail and musical texture point. This is driven home every time I listen to some CDs I know well but are for the first time playing them on my SET. My SE pentode amps pale in the presents of the SET. I may never find it's champion.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 84 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8, 9  Next

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 21 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
DIY Tube Projects :: DIY Tube Amp Kits :: DIY Speaker Projects :: DIY Solid State Projects :: DIY IC / Op-amp Projects :: DIY Phono Projects :: DIY Cable Projects :: Hi-Fi Audio Schematics
© diyAudioProjects.com - Terms of Service - Privacy Policy