|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
97.84.136.207
I know I know,you may have heard it all before but not really because I haven't seen much on comparison of the reissue kt88 vs old genelex kt88. Maybe its because tube dealers that sell both,don't want customers to assume any facsimile of the truth for fear of them expressing anger over a 400 dollar a piece sale price, and then being bested by a tube costing 50 to 70 dollars each.
When I test any new part, or amp, or preamp,anyone that knows me realizes that I don't depend on one thing for judging that component. The brain and human ear can always acclimate our feeling and hearing mechanisms to adjust to imbalance or inefficiencies the equipment or component may have. Being that our ears are not linear,we sometimes are off by a country mile in our subjective listening tests.
Getting to my comparison of the New vs NOS genelex kt88. Everyone knows the sonic virtues of the tubes of yesteryear.What some people don't know is,under biasing a tube and not putting at least 200 hours on a tube is not going to give you a valid interpretation of the tube's overall character.
Last night for the first time,I made a month long comparison between new versus vintage Genelex/Gec.When you under bias the reissue kt88, its going exhibit harshness and transient distortion like you have never heard. When you are generous with the bias,(and when I say generous,I don't mean 60ma),I mean like 80 to 85ma,this tube will rival and and soundly defeat the vintage Gens for sonic richness in the mids and upper high end. Bass is deep and controlled and very energetic,with all the basic flat response and low distortion you would get with a very expensive amp.You need to request the higher transconductance testing ones for audio
.The vintage Gen/Gec were also very detailed and had a tiny bit more focus on the very very top frequencies and I am talking tiny.After that,the reissues just took over. They were more three dimensional,more alive,richer in every part of the spectrum than the vintage was and,it had the sonic character of the vintage TS 6550 black plates.
I honestly think that being New Sensor makes the new tungsol and the new Genelex from original recipe,they were able to combine the sonic virtues of both tube types and came up with the master valve.
I did the test on my grommes 260a amps which have regulated screens and a higher plate voltage.This would give me a honest pentode type evaluation and from the distortion test,it made realize that many people are under running this tube because biasing at 58ma like grommes recommended,made the tube sag and distort in the upper octaves.Instead,I biased at 70ma,and the tubes were the best sounding kt88 I have heard to date,not counting the 6550 tungsol black plates..They are definitely much more alive sounding than their vintage GEN/GECcounterparts.
Just be sure to be generous with the bias because these tubes can take 90ma all day long at typical kt88 voltages.You will not hurt them,trust me. I finally feel that this is the most affluent tube in the industry other than maybe the Shuggie treasure series.I want to hear those again.When I put the sweep generator on from the computer RTA, I couldn't see any peaks or dips in the response. The only thing I had found in new versus old was,the reissues have a little more upper frequency distortion when they are not biased to at least 80 to 85ma.On the grommes and the audio research amps,you typically don't go that high because of the higher B+. Once that bias issue was corrected,I can't stop praising this tube.
If the power supply waveform isn't pretty,neither is the sound in most cases.
Edits: 04/28/10 07/31/10Follow Ups:
KT88 Ratings
Vh Ih VaMax Vg2Max VhkMax PaMax Pg2Max IkMax Cgk Cak Cga
6.3 1.6 800 600 200 42 8 230 16.0 12.0 1.2
KT88 Application Data
Class Va Vg2 Vg1 Ia Ig2 Rk Zout Pout THD Notes
AB1 521 300 2x64 2x1.7 2x460 9000 50 3.0 B+ = 560V
AB1 552 300 -34.0 2x60 2x1.7 4500 100 2.5
AB1 U/L 40% 553 -75.0 2x50 4500 100 2.0
AB1 U/L 40% 453 -59.0 2x50 4000 70 2.0
AB1 U/L 40% 436 2x87 2x600 6000 50 1.5 B+ = 500V
AB1 U/L 40% 328 2x87 2x400 5000 30 1.0 B+ = 375V
AB1 triode 349 2x79 2x525 4000 17 1.5 B+ = 400V
AB1 triode 422 2x94 2x525 4000 31 1.5 B+ = 485VObserve the 87ma at at 500v thru 6k impedance
Observe the 94ma at 485v thru 4k impedance.
Believe me ,the genelex can handle it and Im only saying you can go to 80ma when you have a plate voltage of 440v.
If the power supply waveform isn't pretty,neither is the sound in most cases.
Edits: 07/31/10
can you try to up the bias on the mC60 to see if it sound better ?
Pete
The only way we can do that would be to put a larger value resistor coming off the bias supply to lower it.The bias tracks on the mac and typically its minus 47 at 440v on the plate and then it tracks from there according to change in B+.
You can stick a pot in the bias and try experimenting to see the improvement. I did it with the KT90s in the macs and they were beauty on steroids. They would cook!
If the power supply waveform isn't pretty,neither is the sound in most cases.
Just curious, how high do you bias your NOS Tung Sol 6550s?
Shakey
I bias the nos TS between 70 and 80ma
If the power supply waveform isn't pretty,neither is the sound in most cases.
Mike, I don't remember ever biasing NOS TS 6550's to 70-80mA (each?). Unless you are running very low plate voltages (between 300-350 VDC). I like to bias the TS 6550 to around a max IDLE PLATE DISSIPATION of 25-30 watts per tube. For amps that place 500-550 VDC onto the plates, I start at around 35-40mA per tube (for Class AB1, push-pull amps). Then, slowly tweak the bias upwaard and listen to the amp's response.
Amps that run over 600 VDC onto the plates get KT-88's.
Steve
the original citation 2s run at 460 and they were running 100ma thru each tube. Then,they dropped it to 90ma. The 600v plate,has 300v on the screens and that I would go no more than 62ma.
If the power supply waveform isn't pretty,neither is the sound in most cases.
Mike, 100mA at 600VDC or 60 watts per tube! If I ran that on an Ampeg SVT you potentially get 360 watts! Until it blew sky high. LOL! As it is, the 120 watts or so is SO LOUD it's crazy.
Mike's post isn't clear and the numbers got confused.
The original Citation II biased at 100 ma. and had 450 volts across the tube. That was 45 watts - that was really pushing it, agreed. But back then tubes were relatively cheap and plentiful so it wasn't as big a deal. In spite of that, Stu Hegeman & Harman-Kardon later reduced the bias current spec to 90 ma., which dropped the dissipation to 40.5 watts.
With my kits I recommend 80 to 85 ma. With today's higher line voltages, etc., that gives 37-39 watts dissipation.
No amp I know of ever (deliberately!) ran 600 volts and 100 ma per tube.
One thing Mike pointed out is definitely worth remembering - in a pentode or beam power tube the screen voltage has much more influence on current than the plate voltage does. You can run very high plate voltages with lower screen volts and it'll work well (if it's done right obviously).
Finally, keep in mind that when you measure cathode current you are measuring the sum of the plate current AND the screen current. So 85 ma. cathode current on a Cit II is actually (it varies from tube to tube some) about 78 ma plate current and about 7 ma-ish screen current. That means you have a plate dissipation of 78 ma. x 460 plate volts, or about 36 watts; and you have screen dissipation of 7 ma x 460 screen volts, or about 3.2 watts. Do keep in mind plate and screen current varies some from full load to idle - I chose sort of a "mid-point" above.
A 6550A or KT88 typically have a plate dissipation limit of 42 watts and a screen limit of 8 watts using the Absolute Maximum rating system, the least conservative rating. The next more conservative rating is the "Design Maximum". The Design Maximum rating for the plate is 35 watts.
----No amp I know of ever (deliberately!) ran 600 volts and 100 ma per tube.
Let's see...:) RCA's MI-9355 ran well over that voltage and quite comparable cathode current( no g2 in an 845 ). My 4E27 amps ran near 600V and idle at 110 mA. I'm sure GM100 amps exceed that idle current/plate voltage.
I suspect you speak of 6550/KT88 consumer amps...heh-heh-heh( and in that you'd certainly be correct ).
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.
Hi Mikey,
I agree totally. I have only heard them in a Citation II, but when you put them at 80 - 82 mA they are stunningly good. 75 just sounds a bit lifeless. My Citation II's with the level II+ PS generally run about 452 - 458 volts on the plates depending on the wall and PT variations from amp to amp. At 455 v and 82 mA it is sonic heaven! I had a quad of original GEC KT88 tubes go through here with an amp. They all tested strong. I listened. I preferred the sparkle of the new ones and they were more 3-D too. So I agree with you.
cheers,
Don
Exactly Don
The vintage ones sound ok to but when I was talking my friend Silpachai in Hong Kong,I would say Sil,how do you like the old GEC/Gen KT88s? His response was,Mikey,thats a boring and lifeless sound that puts me to sleep. We talk on skype quite a bit and he is a great violinist and percussionist. He sends me CDs and open reel tapes of his Music and,he is also big into rebuilding and upgrading like we are.
If the power supply waveform isn't pretty,neither is the sound in most cases.
Mike,
When you say bias at 80 mA do you mean for each tube or for a pair? 80 mA for a single tube seems pretty high.
Jim
Each tube is what I mean. Now,not every amp you can bias at 80ma. I have even biased at 85ma but on amps like audio research.I wouldn't go over 62ma. The genelex reissues are very durable and will withstand more punishment than the other new production KT88s.
If the power supply waveform isn't pretty,neither is the sound in most cases.
Hi Mike:
A little surprised by your bias comments---when I had them in my amp I found that at a higher bias they sounded to overly full, bloated and dead on top. I actually ended up running them at a lower bias than the EH KT88 that came with the amp. Next time I put them in I will try biasing them even higher and see what that does to my opinion. I seem to be the only one here not blown away by these tubes.
Mendel
You need at least 200 or 300 hours on the tubes before the higher bias sounds good. It also depends on the amp you put them in.
If the power supply waveform isn't pretty,neither is the sound in most cases.
Mike:
I have about 300 hours on the tubes. Not to beat a dead horse to death............BUT---Rogue Audio recommends biasing KT88 in the Stereo 90 at 40ma per tube. Are you really serious about biasing at 80ma per tube? That is twice the recommended bias. Won't that hurt something/someone? I don't know what voltage is on the plates,but I do seem to be able to run 6L6GC no problem. Is it possible that your 80ma reco is not OK for all amps?
No, you are not alone.
Thanks for the post, Michael. Very informative. You mention being very generous with the bias with respect to the reissue KT-88. I have my quad biased at 50ma and thought that might be pushing it a bit. Won't 80-85ma be pushing this tube too hard, possibly driving it into distortion and lowering it's power? Could that also be putting too much stress on the amp? I love the sound of these tubes at 50ma, but am now curious about adjusting up their bias. But, for me, they are expensive, and I don't want to wear them out prematurely or over-work my amp. Any thoughts?
"I have my quad biased at 50ma and thought that might be pushing it a bit. Won't 80-85ma be pushing this tube too hard,"
It depends. The tube has a maximum current rating WAY over 85 ma., so as long as the dissipation is within reason 80 ma is just fine.
Using a "McShaned" Citation II for example, you'll typically see 455 to 460 volts or so on the plate and just 1 - 1.5 volts on the cathode. So let's say the voltage drop across the tube is 455 volts.
With 80 ma. cathode current we can compute the dissipation wattage using the formula Watts = Current times Voltage. Well, .08 amps (80 ma) times 455 volts equals 36.4 watts. The plate is rated at 42 watts, so we're fine. Also, some of the current you measure at the tube cathode is screen current, not plate current. At idle you can figure on 8 ma. or so as a rule of thumb for screen current. So the plate is dissipating .008 x 455, or 3.6 watts. That's fine too. And if you subtract the screen current from the plate current then the plate dissipation is only 455 volts x (80-8 ma or 72 ma). That computes to just under 33 watts.
"possibly driving it into distortion and lowering it's power?"
Not unless there are other issues with the amp - a tired power supply or ??
" Could that also be putting too much stress on the amp? I love the sound of these tubes at 50ma, but am now curious about adjusting up their bias. But, for me, they are expensive, and I don't want to wear them out prematurely or over-work my amp. Any thoughts?"
There is no hard and fast rule here. As far as tube life goes, there is a finite number of electrons the cathode can emit over its life. So running them harder does wear them a bit sooner. Of course, leaving them in the box reduces wear a lot! So it's a question of what you want - in your car, do you occasionally like/need to accelerate fast? Well, it wears out your car and burns gas faster, so you have to decide if it is worth it.
As far as the amp itself - that's a case by case deal. I think MOST amps work fine with higher bias currents, but it's not a sure thing. Check with the amp maker and ask them about it - you may get a "pat" (no pun intended) answer, but it's worth asking.
Jim
Your calculations on plate dissipation are straight forward.
My PP Cary is modified to triode with a 100-ohm connected across the pins 3 & 4 and the transformer connection to pin 4 removed. So grid 2 is "tied" to the plate via the added R. My concern is whether or not g2 will be able to withstand the high bias discussed herein ?
If not, what would be the "lowered" limit for plate dissipation?
Calculate the current passing through it (G2). It's the same calculation as I just mentioned above.
You can measure the voltage across the 100 Ohm resistor and use Ohm's Law to calculate the current. For instance, you might find .6 volts drop across the 100 Ohm resistor. If you do, then you know the current flow is .6/100, or 6 ma. If the screen is at 450 volts, then you know the screen dissipation is 450 x .006, or 2.7 watts. That's within spec. So is 450 volts - but often tubes have different limits if triode connected so check to be sure. IIRC it's 600 volts on KT-88s that are triode connected.
There is no change to the plate dissipation - calculate it the same way as above.
Okay,
So last night I upped the bias from 50ma to 60ma in my amp. It could be my imagination, but I think it sounds better than before. My amp is a Cayin TA-30 35WPC ultralinear integrated amp. The curcuit is identical to Cayin's current production AT-50. I did a search to determine the plate voltage for this amp and came up with nothing. The stock amp is supplied with a quad of EL34's, but you can also use 6L6 and 6550/KT-88 types, as long as you adjust the bias accordingly. This amp uses very robust power and output transformers and it and the Gold Lions are showing no signs of strain at 60ma. Still, I wonder if you or anyone else here might know what the plate voltage rating is for these amps and what the optimal bias setting would be.
Emailed Paul from Tube Audio Design, who I originally purchased the amp from. He recommended that I don't exceed 50ma in this particular amp as the higher idle current might stress the output transformers.
10 ma per tube could overstress the OPTs?? Wow.
It's their amp and I would hope they know it better than anyone. But boy, that strikes me as saying that the OPTs are not particularly - well, let's just say "robust".
That's my opinion, not a verified fact.
He decribed them as under-spec'd. The amp is supplied with EL-34's running at 35ma. It is capable of running KT88's, but apparently within a lower bias range, with 50ma the recommended max.
Okay, be sure to pay attention to that then!
Will do. This amp has a lot of power, headroom and a wide frequency response for its power rating. I'm thinking it could handle 60ma. But then again, it is ideally designed for EL34's, with the ability to run KT-88's, but within certain parameters. At least that is my understanding. So back down to 50ma. Why invite trouble?
Best Regards,
Pat
Pat
I bias 6L6gcs and EL34s at 50 and 55ma all day long. A genelex kt88 is just getting started at 55ma.
Audio research runs 600 on the plate and a regulated 300 on the screen grid. They say to bias at 50ma but you can go at least another 10 to 15ma.
If the power supply waveform isn't pretty,neither is the sound in most cases.
And since we're on to different tubes and biasing - I have a Quicksilver V4 with Tesla tubes running at the recommended 40 ma. Any opinions on boosting the bias in these particular amps? And yes I'm chicken to do so until I hear something from my fellow tubites out there.
Frank
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: