Can I replace 8417 with 6550 tubes?

[Doug]

Have an old SG systems combo bass amp with two 8417 tubes. I read somewhere you could replace and rebias. Could I convert to 6L6?
Thanks
Doug

[steve]

Have an old SG systems combo bass amp with two 8417 tubes. I read somewhere you could replace and rebias. Could I convert to 6L6?
Thanks
Doug

IIRC, the voltage to drive an 8417's grid is less than that of the 6550, so you can do it, but you may have to adjust the driver tube circuit to get the 6550 to work in the circuit.

Here is a page I found on the Triode Electronics site:

http://www.triodeel.com/8417.htm

HTH,

steve

[Rob Mercure]

Doug,

Are the 8417s defective? I've got a couple of amps that use these bottles and I've found them to be quite long lived. But I've also found that installing a stable screen supply is a must as well as ensuring that the control grid bias supply is also stable and, in my opinion, variable to adjust to individual tubes.

If you do decide to replace the 8417s I like the sound of the USA made "big bottle" 6CA7s more than the 6550 - the 6550 seems a bit "cold" to my ears but many bass players like 'em. But, of course, the "real" 6CA7 may be as hard to find as the 8417. EI is now making a 6CA7 that has gotten some nice reviews but I'm not sure I've seen one that compares it to the USA 6CA7 but instead to it's "cousin" the EL34.

And the 6L6 you asked about may be a possiblility depending on the plate and screen voltages as well as the expected power output. In general you won't be able to get the same output as the 8417s but a few watts difference may not be that noticable.

Anyhoo, with any of these bottles you will, as noted, need to increase your bias supply and also the control grid drive signal.

Rob

[Enzo]

The reason they used those amazing 8417s is that you can get 100 watts from a single pair of them.

[Doug]

Not defective, just thought I'd update...

Thanks guys, the amp seems to be functioning fine, but the output seems pretty low. Sounds good, but not much volume for 2x10s and a 15" (it's an odd combo) can't be putting out 100 watts as it is. I'm wondering if the bias is what it should be, etc... I'll check out the speakers and see what I can learn about the amp before I start refitting tubes. I didn't know these 8417s were so cool.
Thanks
Doug

[Rob Mercure]

Doug,

Yeah the 8417s are neat - while there are several tubes claimed to put out 100W audio from two tubes these are the only one I know that should do it with only about 35 V peak-to-peak drive on the control grids. And this output is easily provided by a split load PI. But it is the very high transconductance - "plate sensitivity" - that allows these bottles to redplate easily as we're only looking at a G1 bias of minus 16-20 volts. So a bad bias supply or just a bit of leakage from the DC blocking gap can initiate meltdown. This is why I recommend making sure that your control grid bias supply is in great shape - with separate bias adjustment for each tube (and I like a 1 ohm, 1% current sense resistor between the cathode and ground and of sufficient wattage to be sturdy - some advocate using this resistor as a fuse but I'd rather use a real fuse if this was my purpose). And having a stable G2 supply is mandatory - you won't get full power if the screen voltage is low and some 8417s have been reported to suffer from screen winding "sag" when run hot which can result in a short to a adjacent support rod (no personal experience on this one).
As Enzo pointed out, these bottles are claimed good for 100W/pair. But so are 6550s and some other tubes. What I was, poorly, expressing is that perception of sound is relative and a smaller amp with more speakers/more efficient speakers can easily be "louder" than a larger amp with less efficient output devices. One can do the dB ratio between, say, 70W (supposedly obtainable from a pair of 6L6GCs) and 100W and note that it's not that much change - the human ear is a notoriously non-linear measuring device coupled to some sophisticated programming that wasn't "designed" for power output comparison <grin>. The oft quoted "rule of thumb" is that you have to increase the output power 10 times to double the perception of sound level. So be aware that your "100W" output may all be there you just can't easily compare the level to other equipment without some "fudge factor" experience. And of course this argues for trying other speakers - especially if you can obtain some with efficiencies around 100 dB/W or better.
If you want to provide us with some output stage information, Ep, Eg1, Eg2, Ip, Ig2, we might be able to figure out if your 8417s are running at some "optimal" level.

Rob

[Don Symes]

Thanks guys, the amp seems to be functioning fine, but the output seems pretty low. Sounds good, but not much volume for 2x10s and a 15" (it's an odd combo) can't be putting out 100 watts as it is.

If you can get the amp connected to a modern, high-efficiency cab you may be suprised what a difference that can make. Trying a known-good 100W amp on those speakers may also be enlightening.

[Steve Conner]

I second that. Amp power doesn't really seem to make much difference, compared to speaker efficiency. I've been amazed (and so have my bandmates and sound guys at venues we played, and not always in a good way...) by how much bass volume I could get out of even a 50 watt tube amp hooked up to a big speaker cabinet.

2x10"s and a 15" is a decent area of coneage, though, so the cabinet you have should be fairly efficient. I'd say, in the light of what other posters said about the ear's log response, if an amp sounds quiet, it must be seriously down on power, as in only putting out 1/5 or 1/10 of its rated output. So definitely check for stuff like bad tubes and blown speakers. (given the rarity of 8417s it's understandable if they never got changed ever.)

I once owned a 4x10" bass cab that I bought used, and it was a good while before I figured out two of the woofers were blown. That would cut output by a factor of 4 not to mention detuning the cabinet, but it was still fairly loud. I sure noticed when I got it fixed though ;-)