Safe alternative to KT66

[tpatton]

With my Quad II amps finally working properly, I tried a quartet of vintage RCA 6L6 tubes.  One amp blew its fuse pretty quickly.  The other didn't blow its fuse, but I wanted to take those tubes out ASAP.  But even with (thin) cotton gloves, I burned my thumb on one tube, now have a blister.  I put all four of those tubes in a toxic waste bin, but would now like to get some spares for when the KT66's go.

Is 6L6GC a better KT66 replacement?  Or is plain 6L6 okay as long as the tubes aren't already somehow bad?  If anyone has experience with 6L6GC, I'd like to hear, especially if they were in Quad II amps.  My thanks in advance to anyone who can inform or advise.

[Henry's Cat]

I've used Sovtek 6L6WGC in my Quad II's for several years now with no problems.  The cathode bias resistor R12's value needs to be raised to 220 ohms though if you change to 6L6's. I've also used 5881's for a while before changing to 6L6's.  Back in the 1990's I tried Golden Dragon 'regular' KT66's but they were a bit gassy as they had a blue glow inside the bottles!  I also tried Golden Dragon Super KT66's which were very good but expensive and I only had one pair.

Re. GZ34's.  They are not listed as equivalents because their specification is different and is actually better. Their heater current is 1.9A, anode voltage rating 550V and cathode current 160mA. The GZ32 specs. are heater current 2.3A, max. anode volts 500 and max. cathode current 125mA.

[tpatton]

Thanks, H's C--this is golden.  I've just ordered from eBay Nixie Store (Russian stock) the following:

6P3S / 6L6 / 6L6GT / 6L6GC / 5881 TUBE. 4 NEW

Possibly with resistor changes, I should be okay with these, right?

I've printed your post and will be guided by it.  Any comments on the above purchase will be welcome.

If you can't find the Nxie Store ad, please let me know and I'll take steps.

[tpatton]

I presently have Mills MR5 180R ohms resistors in as the R12's.  If I change to 6L6GC tubes, H's C says (and I believe it) that I should have 220R ohms instead.  But what if I want to change back?

Here's a way that tears things up minimally.  I put in 220R ohms resistors, for the 6L6GC's, and have on hand 1000R ohms Mills MR5's, and to change back I solder them onto the leads of the 220R ohms ones.

990R ohms would be right on the nose, but I think 1000R ohms should be close enough.

                             220X/(220 + X) = 180, 40X = (220)(180) = 39,600, X = 990

                   Check: (200 x 990)/(200 + 990) = 180

But do the R12's need to be as good (and as expensive) as Mills MRA 5's?  Maybe Mouser has some high wattage resistors that would do just as well.  Advice on that would be welcome.

[tpatton]

I just got this web site from the Russian seller of the tubes I'm buying:

www.hifitubes.n/weblog/wp-content/russian-6p3s-e.pdf

Henry's Cat, could you take a look at this, the column for 6P3S, not the other column, and let m know what you think of using them in Quad II's?  I'm committed to four of them, and might buy another four since they're so cheap and, if other Russian stuff I've bought is any indication, quite good.

Thanks--I'll look forward to hearing from you, or from anyone with a view on this.  I can easily change R12, the most accessible resistor in the amps.

[RayP]

Since the R12 resistor modifier of 1k or 990 will only dissipate 0.68 watts, a 2 or 3 watt resistor should be fine. I would get the 1k over the 990 since there is more choice at 1k and the resulting difference is only 0.328 ohms.

You could consider the Koa Speer 2 or 3 watt 1k carbon film at 24 cents and  32 cents respectively.

http://www.mouser.com/KOA-Speer/Passive-Components/Resistors/Film-Resistors/Carbon-Film-Resistors-Through-Hole/_/N-7gz3yZ1z0zlrr?P=1z0vocwZ1z0vo2aZ1z0vkp6&Keyword=resistors&FS=True&Ntx=mode%20matchall&Ntk=Mouser_Wildcards

If you wanted to stay with wirewound for the 1K then Mouser has a bunch of them rated from 1 watt thru 3 watts.
http://www.mouser.com/Passive-Components/Resistors/Wirewound-Resistors/Wirewound-Resistors-Through-Hole/_/N-7fx9iZscv7?P=1z0vkp6Z1z0x6yoZ1z0vnkoZ1z0vocvZ1z0vl8kZ1z0vocwZ1z0vkbzZ1z0vk6bZ1z0vl8yZ1z0vjzvZ1z0vo2a&Keyword=wire+wound+resistors&FS=True

Mouser also sells a bunch of 7 and 10 watt 220 ohm wire wound resistors (220 ohm dissipates 3.07 watts).
http://www.mouser.com/Passive-Components/Resistors/Wirewound-Resistors/Wirewound-Resistors-Through-Hole/_/N-7fx9iZscv7?P=1z0x8beZ1z0vl80Z1z0vkbnZ1z0vk5nZ1z0seteZ1z0vnpi&Keyword=wire+wound+resistors&FS=True

There's quite a variation in price but you may notice that only one from Ohmite at $1.60 has a 1% tolerance if that is important to you. All the others have 5% tolerance.

I can't give you advice on the sound. I would be more concerned about whether the resistors fit physically in the space available.

Out of interest, I also looked up the Mills resistor prices at Handmade and they seem to be $3.60 (1k@5w) and $4 (220 @ 12w). If they fit, this might be the easiest solution.
http://www.hndme.com/productcart/pc/viewCategories.asp?idCategory=70

ray

[tpatton]

Raay P, thanks for all this good resistor advice, use and buying.  However, the 990 ohms or 1000 ohms ones were only for use in parallel with 220 ohms to get back to 180 ohms if I ever switched back to KT66's.  They were never intended for use by themselves.  So I think highish wattage for R12 is still on.

[RayP]

Yes I understood the 1k were going to be in parallel with the 220. I presume I wasn't clear that in the parallel situation, the 220 ohm resistor will dissipate 3.07 watts and the 1K will dissipate 0.68 watts.

ray

[tpatton]

Ray, I apologize for failing to get your drift.  It occurred to me later, away from the computer,  that your calculation was based on the 990R ohms resistor being in parallel with a 220R ohms resistor, so it would get much less than half of the current.  Ohms Law, if I remember even that properly.

So the 990R ohms resistor (realistically the 1000R ohms resistor) could be lower wattage, which would be convenient since that doubtless also means a smaller physical size.

But if the Russian tubes I'm getting work out, I won't be going back to 180R ohms in any case.

Did I say in this Forum what their specs are?  Here's the place to look, maybe again:

http://www.hifitubes.n/weblog/wp-content/russian-6p3s-e.pdf

I was told to look (only) at the column for 6P3S.

If any Forum member can look at those specs and tell me what they imply as to what I should change in the Quad II circuit--certainly R12, I guess, but maybe something more, that would be most welcome. 

[Henry's Cat]

The link appears to be dead or faulty.  I can't view the PDF file. Could you check the web link.
 
Re. the specs. for GZ32's and GZ34.  The details I found on the internet were operating conditions at given anode voltages, not specifications. On inspecting the Mullard Tube Manual the GZ32 can supply more current but has a lower max. anode voltage than the GZ34. For Quad II purposes it's not really an issue, the GZ34 will still be fine for the Quad II.

[RayP]

The link for the datasheet should be http://www.hifitubes.nl/weblog/wp-content/russian-6p3s-e.pdf

You may already have seen these but take a look at
http://www.diyaudio.com/forums/tubes-valves/163839-quad-ii-can-6p3s-e-gu50-used-kt66-direct-replacement.html

also
http://ampgarage.com/forum/viewtopic.php?t=12456&highlight=&sid=88bdfce552d5c0200d8166a8d9ad3b47 - down the page where it says
'Next time try to find 6P3S-E version, E stands for military use, it's long life tube, and better than standard. Remember that normal 6P3S is like 6L6, while 6P3S-E is 6L6GC, in most datasheets you will find max V for both about 250V, but many people use them with much higher values (good replacement for Fender amps). In amp that I've got on desk currently is 430V on anodes, and I used 6P3S-E.'

Since the data sheet link you supplied shows that the curves are the same for both the E version and the standard version, I suspect your standard version will be fine. I see it has a higher max anode voltage of 375 compared to the 250 of the E version. I suspect that the E would be fine at 375 as well but as a military version it would be warranted for use only to 250.

I took a look at a Quad II schematic at
http://www.keith-snook.info/Schematics/QUAD%20II%20Schematic.pdf

It looks like each KT66 is dissipating (340 - 26)*.065 = 20 watts which is the max for both the standard 6P3S and E version according to the datasheet. If they are cheap, then you can afford to replace them occasionally.

As to changing R12, I seem to remember reading plenty of advice that if your voltages are within 10% of what you are aiming for, you will generally be ok. Try the 220 and see if it works for you. If not, try the 180. Just make sure that your anode dissipation does not exceed 20 watts.

I hope this helps.

ray

[tpatton]

tThanks very much, Ray, for this final reassurance.  I will go to the 220R phms, and have a way, already discussed, to get back to 180, either for the incoming Russian tubes or for the old KT66's--but I doubt if I'll be going back to them for the foreseeable future, with eight of the Russian tubes on their way.

I'll call Mouser today and order some resistors.

I'm so conditioned to scary failures that I was timid about even turning the Quads on this weekend, but finally did yesterday, and they played flawlessly, no blown fuses or melting down tubes or anything, and they even sounded quite good, and can sensibly alternate with my great-sounding Bottlehead amps.

I'm going to see if I can recover my previous discovery that the formula for parallel resistors comes right out of Ohms Law, even to the point of getting the formula for three or more resistors in parallel, which is a lot more complicated than the formula for two.  I remember doing this a few years ago, but my braid may have gone a long way downhill from then.

Thanks again for your words on this--they very definitely did help.

[2wo]

"they played flawlessly, no blown fuses or melting down tubes or anything, "

Oh, what fun is that?

[JC]

tpatton, you might find this useful:

http://www.1728.org/resistrs.htm

Way back in 1980, I wrote a program in BASIC for my TRS-80 to do this for me.  I remember inserting "wait states", because it was so quick about it!

[tpatton]

Jim, that's really neat!  Haven't tried it out yet, but I will.  The initial equation was the one I'd arrived at before, from a physics book discussion of Ohms Law and its test problems.  I worked out something fully general, but who'd ever want to have more than 10?  But it's interesting (at least to me) to see how the case of two doesn't even suggest the complexity of the general case.