Subbing 6SL7's in 12AX7 circuits

[EtherealWidow]

Let me start off by saying that I realize that the 6SL7 is not a direct replacement for 12AX7 tubes. Aside from the socket difference, there is also plate voltage, filament voltage, etc. What I'm wondering is if you guys have experience in taking circuits that were designed for mini 9 pins with octals. For example, Expresses, or even 18 watt Marshalls or old Tweed amps. How do the octals stack up against the mini 9 pins? Do you change the coupling cap values? How do you have the octals wired up? (grid leak?)

The reason I ask is cause I saw one of you post a link to BC Audio. Dude's got some serious mojo.

[Cantplay]

If you're going to resocket and rewire, why not go to a less expensive option?

7N7 is the loctal version of 6SN7.

20EZ7 and 6AX7 are similar but diff heater.

John

[martin manning]

These two will will operate quite well with all of the same circuit values. Using typical 100k Ra and 1k5 Rk, the DC operating point looks well centered for both. The 12AX7 will have a bit more gain (34 vs. 32 dB at midrange), and the LF roll-off is nearly identical. The -3 dB HF roll-off is higher for the 12AX7 due to its lower input capacitance. Some tweaks could be made to compensate for that if the 6SL7 loses too much HF.

[EtherealWidow]

If you're going to resocket and rewire, why not go to a less expensive option?

7N7 is the loctal version of 6SN7.

20EZ7 and 6AX7 are similar but diff heater.

John

I wasn't going to rewire an amp. I was going to build one based off a 12AX7 circuit, but with the octals. Sorry. Should have been more clear. The 7N7's might come in handy, though they're a bit hard to find. Thank you!

These two will will operate quite well with all of the same circuit values. Using typical 100k Ra and 1k5 Rk, the DC operating point looks well centered for both. The 12AX7 will have a bit more gain (34 vs. 32 dB at midrange), and the LF roll-off is nearly identical. The -3 dB HF roll-off is higher for the 12AX7 due to its lower input capacitance. Some tweaks could be made to compensate for that if the 6SL7 loses too much HF.

Awesome. Not too worried about the gain. I know they only have a mu of 70, but hey, I'll be cascading them anyway. No biggie. I don't even think I'll be overly concerned with the top rolloff either. I play pretty bright, not really by preference. Would it sound weird to cascade grid leaked octals into each other? I've just never seen a schematic with that kind of setup. And I hear they can be a bit noisy with heater hum. What do I do about that? I know that BC Audio uses a DC power supply.

[David Root]

Years ago I built an amp with all 6SL7 pre and PI tubes. It morphed thru a few different incarnations and ended up as a Dumbleized 5F6A tweed Bassman running 100W off a pair of 8417.

The 6SL7 seems to have more space, openness than the 12AX7,

BUT one thing I learned early on is that if you are going to put the 6SL7 in hi-gain applications be sure to use DC on the filaments, otherwise they will hummmm like crazy because the filament cannot be paralleled, unlike the 12A*7 Novals. I think this likely applies to all octal pre-triodes. Don't know about pentode pre-tubes though, maybe not.

[EtherealWidow]

Years ago I built an amp with all 6SL7 pre and PI tubes. It morphed thru a few different incarnations and ended up as a Dumbleized 5F6A tweed Bassman running 100W off a pair of 8417.

The 6SL7 seems to have more space, openness than the 12AX7,

BUT one thing I learned early on is that if you are going to put the 6SL7 in hi-gain applications be sure to use DC on the filaments, otherwise they will hummmm like crazy because the filament cannot be paralleled, unlike the 12A*7 Novals. I think this likely applies to all octal pre-triodes. Don't know about pentode pre-tubes though, maybe not.

Sounds awesome. I'm pumped. What did you use for your DC supply?

[passfan]

I know a lot of people claim problems with them but I use them all the time simply rewiring the sockets and using the same values and they sing. Normally in the rocket but recently an Express and people around hear are fighting over it. The darn things been sold 3 times in a month and a half and the price keeps going up. That said, I do have a stash of vintage squirrled away that sound better than the newer ones. You do find one every now and then that makes weird noises and there's no fixing it. You just have to get another.

[David Root]

Ethereal Widow, for the 6V3 DC power supply I used the heater winding into a 25A 50V bridge rectifier followed by a 15,000uF 16V cap on each side, with a 1.5mH 5A choke (Hammond 156B) on one side and a 0.3R 5W sandbox on the other side to get the voltage in spec. This fed all octal pre-s including the PI. The 8417s just got straight 6V3 AC.

Passfan, I got one of those 6SL7s that weirded out on me. Too bad, it was a '51 Ken-Rad, nice tube.

[Cliff Schecht]

When I first built my 5C3 with 6SL7's a few years ago, I couldn't get it quiet without DC heaters. But that was a sloppy PTP build that I was never very proud of or would trust. I rebuilt this thing with an actual turret board and was much more careful with the layout and even with running of of AC heaters, this build is quieter than the DC heater-based earlier build. There is a lot of gain in that amp and it never gets any noisier than a cranked AX7 based tweed Fender amp.

The biggest trick is finding those 6SL7's that really are low noise. I bought a shit ton of them over the past few years only to find a handful (usually military grade versions or special sorted ones) that are low noise enough to use. I use the noisier ones later on in amps (second stage or PI) and save the holy ones for V1 applications. I love my 5C3 to death but I don't use it often just because of how hard good 6SL7's are to find.

The specially sorted ones usually have a small white dot somewhere near the base. These are the most consistently low noise 6SL7's I've found.

[Aurora]

Most classical oscilloscopes and various tube based instrumentation amps used 7-pin single triodes. Anyone here used some of that stuff?
I happen to have a box-load of various types............

[EtherealWidow]

David Root, thank you for the DC supply. I found some more simplified versions of it online, why did you choose to add more elements to yours?

Cliff Schecht, it seems though that most people who use the 6SL7's suffer from AC filament hum. Although layout is crucial, I think the DC supply would help a lot. Plus, I don't have the money to buy a bunch of octals and then pick my favorite. :/

[EtherealWidow]

David Root, would this be the schematic of the supply you're talking about? Disregard the grid leak cascode... https://www.dropbox.com/s/sar0o5nakxr5o ... ascode.jpg

[Blackburn]

I've been a little curious about DC filaments for a bit and you guys are shedding a good deal of light on the subject. Do you guys believe it's worth the expense of extra space and parts or is it more of a last resort for noisy tubes? Is it difficult getting the voltage exact after rectification? And it's interesting about Cliff mentioning a 5C3. I built one too, completely stock and PTP, and I don't recall excessive noise compared to non octal pres of my other builds. I'm curious about this, so I'll play that one tomorrow and see for myself. Very interesting thread.

EDIT: Sorry, forgot Cliff mentioned 6SL7s rather than stock 6SC7s.

[Cliff Schecht]

Something else to consider is that it is usually harder on the heater elements to run them on DC if they are not specifically designed to do so. The elements develop "hot spots" that weaken the heater over time and cause them to eventually fail, and much sooner than if ran off of AC.

I don't think using DC heaters is wrong by any means, but I'm yet to build an amp that absolutely must have them (aside from the one which I ripped apart). In anything but ultra-high gain situations, good layout will quiet down just about any heater hum. Or at least this has been my experience (assuming all e-caps are new/up to snuff and tubes test strong).

It's much trickier (especially considering parts count) to run an amp with DC heaters if you don't have anything over a 6.3V winding. It's usually a stretch to use linear regulators because of the dropout voltage (also you have to consider the diode forward voltage drop and any other losses). It's done all of the time really but I don't like how close to the margins you have to run this setup to get 6.3V DC from 6.3V AC.

The even more complex but sexier solution would be an AC to DC buck/boost SMPS that keeps the preamp tubes right on 6.3V at all times irregardless of input voltage. I wouldn't feel so bad about using my good old stock tubes if they were spot on 6.3V at all times. FWIW I've been contemplating the design for a small but versatile SMPS that can be dropped into any amplifier preamp to run the heaters on DC. Any interest here?

[EtherealWidow]

I hear that filaments ( and maybe this is just for 12ax7's) can be run off 6.3v +/- 10%. 5.7v-6.9v would be fine I imagine. Correct me if I'm way off base.