• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

EL509 Alternatives for this amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Well, I just completed my latest amp and am fed up with the quality of available EL509s. Arc overs, open filaments, imbalances and cracked base glass and I have less than an hour on this amp.

Is there an alternative tube I could run, preferably something octal? I am not against changing component values/voltages to swing this as I'd really like to get it going. Attached is the schematic- the Hammond 150 watt xformers are fed with 700v, 150v on the grids. R20/21 are 1k, R18/19 are 10ohm 5w. I know few tubes will probably match performance and I don't necessarily need as high a power as this amp will run, but please, anything but Magnoval :nod:

What do you think? Any options? I suppose 6L6GCs are out?

Thanks!
Cory
 

Attachments

  • amp.jpg
    amp.jpg
    36.4 KB · Views: 1,608
diyAudio Senior Member
Joined 2002
Hi,

Is there an alternative tube I could run, preferably something octal?

KT90.

Allow me to say that you must be the first person I hear of having such trouble with EL509 which is known as a very rugged sweep tube...this really surprises me.

Do you have an idea about the manufacturer, origin of the misbehaving EL509s?

Cheers,;)
 
How About...

fdegrove said:
Hi,



KT90.

Allow me to say that you must be the first person I hear of having such trouble with EL509 which is known as a very rugged sweep tube...this really surprises me.

Do you have an idea about the manufacturer, origin of the misbehaving EL509s?

Cheers,;)

12E1 ? Cheap and plentiful newold stock . Octal base , topcap and 35w dissipation . I tested some KT90 a while ago and strangely the things didn't smell very nice when they got hot poooooooeyyy !

ps Cadman : What brand of EL509 did you use ?

316a
 
diyAudio Senior Member
Joined 2002
Hi,

...The stories I read?:eek:

Just hope you fellas buy your valves wisely, ie. with some return warranty???

Don't buy some new EI Yugo EL509s but opt for the Russian ones.

The KT90 is of course nothing else but an octal version of the EL509 wich had its origin in former Yugoslavia.

Rumour has it that the post war EI QC sucks, so get NOS while stock lasts.

12E1 ? Cheap and plentiful newold stock . Octal base , topcap and 35w dissipation

Sounds fine to me, if you're in the U.K...

Cheers,;)
 
Not forgetting...

fdegrove said:
Hi,



Because they don't have a Martin Billington over there to push valve prices to stupid limits, right?:rolleyes:

Useable for audio, useable for audio, useable for audio, .......................yuk.

Cheers,;)

...Langrex with the incredible teetering raft of 12E1 ! Mr Billington is a nice chap but he relies on his consultant to decide if the valves are useful or not , saleable is another matter . You're right though , some prices are a little high these days but fine if you're spending 400 quid.... I've heard he's got the most 6AL5 and EF91 in the whole world . Potentially sitting on the shelf and no-where near audio ;) Going back to Ei , I think their stuff is to be avoided these days , smelly KT90 or not . For instance a good way I find of telling if their 12AX7 is going to be a crock of doo-doo is to shake them . Any 'tinkling' rattle then chuck 'em as they're probably going to be microphonic . On the tester most are ****ed-as-a-fart between sections . The few good ones on the other hand are excellent , it's a pity Ei can't sort it out . After all they have the sealex machines from Philips and the tooling . If only they'd take their time to get it right !

316a
 
EI Tubes

Cory

I'm sitting on 10) EL34 and 10) KT90 tubes that were handpicked at the EI factory this March. I guess I should mention the 20) or so 12AX7's also. Anyway these were tested at the factory by a friend that used to live there. They have not been matched however. It sounds like you are north of me up around Waterloo. I'm near Cedar Rapids. You might want to give these a try.

Joe
 
Joe, good to hear from a fellow enthusiast so close to home, we must be less than an hour away-

The 509s are actually Svets from ebay but there's a thread right now at the transcendent board about the bases cracking. Turns out I'm not alone.

The EL34 sounds like what I'd like to run, and it turns out I have a stash of 6CA7 Hitachis and a pair of NOS RCAs so I'm set.

Question is, what changes to the ckt would you guys make to run these? Do you think 150v on G2 is too low? I'd like to make some power with this one.

Any suggestions?
 
Yeah, if you want to change over to EL34s or the like, you'll need a lot more screen voltage. My ST-70-based pentode amp uses about 480V on the screens. Curcio's used 390. Audio Research typically used about 400. That's not necessary for high-perveance tubes like sweeps; in fact, most of them have pretty low max screen voltage ratings.
 
Been following some of these threads and will attemp to make my first post here. :)

I think you can use EL34s if you up the screens to 450-500v as SY said. This should keep your bias voltages within the range already built into the PS. Your cathode resistors will probably show slightly higher voltage (.1V or so) because you'll be pulling a little more idle current, but this won't matter.

It may be that the 700V is close to the max for EL34s, I can't quite remember this number.

And the power will go down. Since the max Ik for the EL34 is around 150mA, your power will go down below 100W but might max out above 70W. Parallel a couple of EL34s and you've got it made.

Also, your volume will have less range because you won't need to supply as much grid voltage to max the power. That is, there is probably too much gain in the front end for EL34s in this configuration.

I am curious to know from the other far better tube jockeys than I, why the follower on the first stage? It seems like you could drive the phase splitter directly from the first stage no matter what tube you were using.
 
Should have thought a little more before I hit submit.

I quickly looked on the hammond site to see what the primary impedance might be for those transformers. I guessed at about 2k ct based on the transformers that were listed at 120W.

So, with EL34 and this low of a load, the power should drop to around 40W.
 
Well, looks like the EL34 has a max plate voltage of 800v so I think I'll be alright running class B 700v. For screen, I'll shoot for 400v, Bias V will end up close to what I've got now.

What's odd is this pair of Heathkit 55w W-7A monoblocks I've got here. They've got 495v on the plates and 498v! on the screens. (which exceeds my databook max values) Sure sounds sweet though.

Below is my regulated 150v screen supply- Not having much experience with regulation ckts, what's my best bet of altering this for 400v output at the cathode? It's fed 700v on the plate. Tube is 6BM8. Would toying with the feedback resistors R35 and R37 do? Ignore everything after R38, it's impressed DC for the AC filaments. I know that R39 is a simple load to keep the tube in its conducting region.

Thanks!
Cory
 

Attachments

  • scrnreg.jpg
    scrnreg.jpg
    18.6 KB · Views: 726
diyAudio Senior Member
Joined 2002
Hi,

Tube is 6BM8. Would toying with the feedback resistors R35 and R37 do? Ignore everything after R38, it's impressed DC for the AC filaments. I know that R39 is a simple load to keep the tube in its conducting region.

Basically, yes...making R37 partially adjustable would give you some range to toy with.

Better would be a complete redesign of the regulator altogether IMHO.

Cheers,;)
 
Cadman, without knowing what your component values are now, if you're willing to simply change component values, this regulator should produce ~480V with a total idle current of about 6mA not including the EL34 screen current.

If you want a variable supply, replace the 165k resistor with a 150K resistor in series with a 50k pot. This will adjust from about 453V to about 500V.

An externally hosted image should be here but it was not working when we last tested it.


This was a quick redraw, but I believe it will work correctly.
 
One other thought is to try the EL509 MKII. These are octal based EL509s. If the problem is noval pin outs, this might solve it.

I only suggest this because you're going to lose a tremendous amount of power if you use EL34s. If the octal EL509s work, you can avoid rewiring the screen supply and you'll have to replace the novals with octals in either case.

You can check out the MKIIs at Tubebuider. MHO. :)
 
diyAudio Senior Member
Joined 2002
BACK TO SQUARE ONE....

Hi,

One other thought is to try the EL509 MKII. These are octal based EL509s. If the problem is noval pin outs, this might solve it.

Not many people actually know this but the KT90 is nothing more than a rebased EL509 (EI).

The EL509 MkII is the same thing but they're not allow to call it a KT90...

I only suggest this because you're going to lose a tremendous amount of power if you use EL34s.

While I like the way the EL509s sound, I've never been a great fan of the European EL34s, the U.S. 6CA7s usually sounding much better in the bass department, YMMV.

Cheers,;)
 

PRR

Member
Joined 2003
Paid Member
Alex- good to see you here.

Cadman --

> the quality of available EL509s. Arc overs, open filaments, imbalances and cracked base glass

Get better 509s? Open-filament, if you are not doing something very odd, is so unlikely that the vendor should take it back.

Arc-over.... welll, running 700V B+, arc-over is a risk. There are reasons why hi-fi amps rarely run much over 500V B+. Some of the PA amps using higher voltages were known as tube-killers. Even when the nominal plate voltage rating allows 700V, transformer action, clipping, reactive loads, and other oddness give much more trouble at higher voltage.

I am trying to figure what you expect to get here. 700V B+, fix-bias, and apparently a 1.9KΩ plate-to-plate impedance: that roughs out to something like 360 Watts RMS! That is a fork of a lot of power to expect from any pair of octal or noval tubes. Even if you do not actually ask 300++Watts of it, the idle voltage and impedance puts a lot of strain on the tubes.

There are few small (not Transmitter) tubes that will stand 700V. Oddly the original EL34 was rated 800V, though some folks think that's a bit high, and the plate power rating won't allow large current or power. With a perfectly regulated plate supply and a precision resistor load, the EL34 claimed 100 Watts RMS; but those conditions are absurd except maybe in some specific radio transmitter.

The 6550 and the 8417 will usually stand 700V, though this is over their rating. Available 8417 are expensive and fragile; while I loved and lived by 8417 in a past life, I would no longer use one. 6550 and kin are available and mostly quite robust. A pair of 6550 working at 600+ volts will give up to 100 Watts, but the load resistance should be more like 4K or 6K, not 2K. Forced to use a 2K transformer, I would think about a quad of 6550, though if your chassis is built there may not be space.

6L6GC is rated 450V Plate. We know that guitarists often run them at far higher voltage, and some as high as 700V idle (sagging to much lower on power-chords). You could find out what brand and batch works good in high-power guitar amps, then spark-test a few to find some that live with 700V on their plates. Looking into a 2K load, I really think you need four or six 6L6GC to handle the current.

If you can lose the 700V B+ supply: still the 2K transformer winding tends to mandate more than a pair of tubes. Few tubes have large-signal effective resistance much less than 2K, and the 2K P-P winding presents 500Ω to each side. With triodes, you would want Rp to be around 250 ohms; pentodes do better for their size but still there are no standard pentodes that fat.

Quick answer: put an 8Ω speaker on the "4Ω" winding so the effective load is 4K. Bass response and power will suffer a little compared to the tranny specs, but it will work. Run 6550 at 700V B+ and about 300V on the screens. Bias the G1 at about -40V (I don't have a 6550 sheet handy). The cathode current must be held to 50mA or less at idle (a wee bit more if you want to push the ability of an original TungSol-made 6550). It will run far up on the B side of Class AB, often not the happiest sound. Without heavy feedback it will jump and bark as the gain changes with signal level. With heavy feedback: why run such high voltage and low impedance to get the same sound as any other high feedback amp?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.