6P14P-K

[mozz]

Building a scratch amp from parts laying around. Power trans is kind of hot for EL84. I have some of these Russian 6P14P-K tubes.(marked OTK cccp) Have the output section built, cathode bias push pull, 8.4k output trans. I wanted to measure plate voltage and screens and figure out dissipation. Plate voltage is 410v, screens at 365v, 13.2v at 250 ohm cathode resistor. Rob robb's calculator tells me for a 12w tube, i'm at 82.5%, for a 14w tube, right at 70%. I think i will be ok, just scared about running these tubes at over 400v.

Power supply is a stock HP section(1962 signal gen) , plate 5Y3-10uf. Screens 5H(250ohm)-20uf., 10k-40uf. 100 ohm. Wanted to drop the screens for some sort of piece of mind (45v lower than plate).

Anyone push these that high? Results?

[Justin Thomas]

Paging Chuck H!??!?!!??

If anybody knows I bet it's Chuck H...

Wish we could tag each other here so.e days!

Justin

[nosaj]

If anybody knows I bet it's Chuck H...

Wish we could tag each other here so.e days!

Justin

Read Russian? Info layout is similar so you maybe able to cull what you need here.

nosaj
http://www.siberian-shop.com/manuals/6p14p.pdf

[mozz]

From what i can see, most of the Russian tubes are 14w. The -ev -eb are supposed to be rated at higher voltage too. Since these are -k, i'm pretty sure they would be also higher rated than a regular 6p14p. Maybe tomorrow i will crank the variac up to 120v and let it sit idle while watching the mains current draw. If it passes that, then again with signal.

[pdf64]

If we’re happy that 6L6GC have a 30 watt plate (design max), then el84 plates are beefier than 12 watts (design centre).
As a reference point, 6V6 are 12 watts design centre, 14 watts design max.
But then el84 and el34 pentodes seem less resilient to over dissipation than their beam tetrode equivalents.

Whatever, the HT seems somewhat excessive for cathode bias, which requires operation close to class A for it to be worthwhile. If the cathode Vdc at obvious clipping is up near or above 150% of its idle value, then I find that fixed bias (applied either at cathode or grids) is the best option.

[Chuck H]

I'm pretty sure they can take it. I gave a pair to a friend that's running them in a Mesa Subway amp with 400+Vp and not much less on the screens, BUT... That amp isn't driving the grids all that damned hard. If you want to drive the grids into mush (like me) you'll do well to reduce the voltage a bit. The tubes will sound better for it too (edit: this will also mitigate overdissipation of the screen grids which can happen with high Vp/Vg2 and hard grid drive). I'm using that exact tube in my own combo at 365Vp and 335Vg2. When I first built this amp it had about 405Vp and sounded ballsy for small bottles, but didn't get that smooth, gritty, intermodulation laced OD that el84's are known for. So I stuck a 150ohm resistor right in front of the rectifier and WA LA! Sounded just right. I have them on a hot cathode bias with a 160R cathode resistor and that's what I'd suggest. IMHO the max current calculations for cathode bias from Weber and Rob's site are usually too low. Consider that Vox, Orange, Matchless and others run el84's at over 14W dissipation and the forum isn't flooded with repair issues on those amps. (edit: and keep in mind that that those services are calculating total tube dissipation and not subtracting screen current or accounting for cathode bias shift self regulating with the current of signal conduction)

If you're using a diode rectifier you can try adding the "sag" resistor and drop some volts. (edit: that is, if you're using a tube rectifier then adding the additional resistor may cause excessive sag) Then go to a 160R or 180R cathode resistor. The extra idle current will also bring HV down a bit.

FWIW those tubes you have are the very best I've tried so far in my own combo. They sound as good as any other and they're holding up A LOT better than other brands/types against becoming all rattly and microphonic. (edit: the way I see it a 10,000 hour tube that isn't "ruggedized" against microphonics is a waste of time and money with small bottles in a combo amp because they'll rattle apart in fewer hours than their ruggedized counterparts. The "k" suffix is a 3000hr tube, but is the best and longest lasting version of this tube in combo amps for this reason IMHE.)

[Dave H]

Here's an English data sheet. It says Pa max for 6P14P is 14W and it's also rated at a higher max voltage than EL84.

6p14per.pdf

[mozz]

Pretty sure Rob's calculator subtracts 5.5% for screen current. Let it sit idle last night for 1/2 hour, wall voltage 122v, plate 408v screen 365v, 13.5v on 250 ohm cathode resistor. Drawing 60 watts. Gonna finish building the rest of the 18watt preamp and PI then see what it does with signal.

5y3 in there now. What output trans are you running?

[Chuck H]

I'm using a Hammond 1608. I like it better than the proprietary Heyboers that were made for this design. I did have to use a low value shunt filter (1500pf cap in series with a 10k resistor) across the primary to get some ringing and a significant HF spike under control This could well due to the more hifi nature of this OT since I've seen a lot of reports that many el84 amps do this and some OT's seem to mitigate it better than other solutions. That said, I still like the 1608.

[mozz]

Dumb me. I ran a 10k node for the screens after the choke. Way too much, screen voltage was dropping terrible when signal was applied. Went to 1k and things and output power was much improved. Then i went just the choke, 10k to PI, 10k to V1. 10 watts clean, 20 watts clipped maxed. 8.4k into 8 ohms. Output plates are at 402 and screens about the same. Output cathode voltage has gone up but i'm under 80%. I do think i will have to do the zener on the center tap. Need a speaker and a guitar, and wife to go shopping.

Other choice, since there's a fairly large choke, is to lower the first cap or even go to a choke input. How much that will drop, or how that will effect the sound i do not know. More experimenting.

I may have done that once before and went with a old oil filled cap of 1uf (or less)at 600v.

[Chuck H]

Dumb me. I ran a 10k node for the screens after the choke. Way too much, screen voltage was dropping terrible when signal was applied. Went to 1k and things and output power was much improved. Then i went just the choke, 10k to PI, 10k to V1. 10 watts clean, 20 watts clipped maxed. 8.4k into 8 ohms. Output plates are at 402 and screens about the same. Output cathode voltage has gone up but i'm under 80%. I do think i will have to do the zener on the center tap. Need a speaker and a guitar, and wife to go shopping.[ATTACH=CONFIG]54179[/ATTACH]

Other choice, since there's a fairly large choke, is to lower the first cap or even go to a choke input. How much that will drop, or how that will effect the sound i do not know. More experimenting.

I may have done that once before and went with a old oil filled cap of 1uf (or less)at 600v.

Well it IS a guitar amp. Some "sag" is acceptable. I probably wouldn't use a choke only because it keeps the screens at a higher voltage. You can often get away with higher than ideal plate voltages as long as you keep the screens closer to spec. Also... The choke needs to be of a high enough inductance to present a high impedance at any HF frequency that could spike at audio levels. Otherwise you risk having a lower impedance to that HF spike at the screens than you have at the plates because the OT will have a relatively high impedance at HF (partly due to real speaker impedances). This can put the tubes in jeopardy if you plan to overdrive them much because the screens will be overdissipating on these HF spikes. This is a common problem with el84 type amps. Especially because they're typically using an 8k+ plate to plate impedance that is actually MUCH higher at spike frequencies. This is really only a problem that presents itself under heavy over drive conditions. And it's under those conditions that I made my observations.

Just something to look out for and I have come to this observation through repeated bench testing. My own amp uses a 2.2k resistor between the plates and screens and individual 470R screen grid resistors. Some people here have recommended 1k screen grid resistors, but the 470R value sounds much better in my amp for dynamic response and perceived power and volume. Since I haven't measured any (overly) dangerous operating conditions with the 470R value I've stayed with them for the tonal benefit.

JM2C.

[tubeswell]

6P14P are good well over 300V (and are long-life tubes as long as you keep the screens under 300), and are 14W tubes.

The rutubes poster (#4) on this thread explains the longevity ratings: https://www.diyaudio.com/forums/tube...14p-tubes.html

Not sure how heavily Donald Trump has cracked down on importing from Russia. rutubes and siberian tubes are good people to deal with, but the shipping takes quite a while longer than usual. Also, don't just go by what is on their websites, these can be out of date - email them directly, they often have piles of tubes that their website says is 'out of stock'

For estimating screen current, the Ia:Ig2 ratio is about 9:1 (8.5:1 to 9.8:1) for most of the typical running situations charted on the Reflektor data sheet. https://frank.pocnet.net/sheets/113/6/6P14P.pdf