Octal equivalent to 5751? (and now 5C3 thread)

[Cliff Schecht]

I whipped out my 5C3 on Saturday so a friend could try it with some slide. I used to like to play this amp dimed but I quickly noticed how gainy the amp is with the 6SL7's I have in there. It was distorting with the volume set at like 9:00-10:00 which seems a bit excessive. FWIW the sound was not bad, it doesn't fart out or get too muddy, just a LOT of gain on tap which makes it hard to dial back.

I did the diode clamping mod on the output stage which fixed the cold bias shifting issues at high gain. Again I think the amp sounds fine, I just want to knock the gain back some. I don't like 6SN7's in this amp though because it looses the ability to get much distortion at all. I'd like to find something that is an octal equivalent to a 5751 (like how the 6SL7 is a 12AX7) without having to drop in an 8 to 9 pin adapter. Are there any tubes I can try that match what I'm talking about?

[M Fowler]

Maybe the 6SC7?

[Cliff Schecht]

That's darn near exactly the same tube as a 6SL7 with a different pinout. Not an option because the 6SC7 costs way too much (many old amps used these so they're more scarce than 6SL7's) and I don't want to rewire sockets or shove adapters in this amp.

My other thoughts on the overtly gainy nature of this amp is that I could shove in some long plate 6SL7's to maybe smooth things out a bit. The 6SL7's I have in there now are (I think) the shortest, smallest plates of any 6SL7 I've ever seen. Maybe this is akin to how a short plate 12AX7 tends to be much more gainy than a long plate..

Really I'm just wondering what others would do. Should I modify the amp to accept "modern" signal levels? Change the grid leaked input stage to something more standard? I don't really want to deviate too far from the original circuit..

[Phil_S]

Cliff,
If you look at the spec sheets, the 6SL7 is pretty much the 5751 in an octal bottle. It is much closer to the 5751 than 12AX7. 6SC7 is even closer to 5751.

6SL7, Mu = 70.4 and Ra = 44K
12AX7, Mu = 100 and Ra = 62.5K
5751, Mu = 69.6, Ra = 58K
6SC7, Mu = 70.2, Ra = 53K

Are you using old stock USA made 6SL7GT? This is the thing to get, not new production, IMHO.

If the amp is so "gainy" as you put it, maybe you should think about biasing the 6SL7 for less gain. This would probably be the simplest thing to do. You might experiment with pot on the cathode or with the value of the bypass cap if you have one. A small tweak might be all it needs.

[M Fowler]

The new production Sovtek 6SC7 is $16.95 Triodestore

GE NOS metal is $23.00 google.

[Cliff Schecht]

Guess I should have checked the data before making that statement above, I thought both had a mu closer to 100. Looks like both are already closer to a 5751 than AX7. The 6SN7 has a measly mu of 20 though and as I said before I feel like the amp loses too much of what makes it fun with these put in there. I'd like something closer to 45-50 mu just for the first stage. Another thought is that I could try a 6SN7 in the first stage and jump the two channels..

I've got more than enough 6SL7's to roll some through the amp and see if I can increase the usable range before distortion. From my understanding even the original 5C3's are very easy to distort though; the grid leaked input stage distorts itself with hot pickups. I guess the other thing I was doing that was different was I was running two Jensen speakers from the 50's that add much more distortion than I'm used to. I mostly run amps through a 1x12 alnico SRO which is hard to distort with a 20W amp.

Anybody else played with this circuit? Doesn't seem like a popular build but it's a fun little amp that can get a pretty mean sounding distortion when cranked. Sort of a unique design compared to later Deluxe designs (and Fender amps in general) but still has that nice characteristic old Fender midrange-heavy response.

[M Fowler]

Here is a difference Deluxe 5C3 schematic to look over.

[labb]

Cliff, I have built two 5B3's. The first one really came out good. So good in fact that a pro player played it for about 6 months and then bought it. Said it was the best sounding amp he has come across in years.(Plays Jazz and contemporary).It has NOS 6SC7's and EH 6V6's. Iron is Magnetic Components 40-18027 PT and 40-18022 OT. Uses a Sovtek 5Y3 rectifier putting 340 VDC to the power tubes. 91 and 92 to the anodes of V1 and 161/164 to the PI.

The second one I built the same but used new Sovtek 6SL7's( also tried a Tung-Sol). Same iron but I had to use NOS 5Y3 to get down to the same anode voltages. Lots of problems with this build. All probably my fault but I can't get it sorted out. The amp has high level of noise(static like) and oscillates really bad. The oscillation comes in with the vol above 5 on either channel. I have checked solder joints and chop sticked until I gave up. Replaced the 6SL7's with NOS 6SC7's and it does the same thing.

One thing that I notice is that both of these amps sound better at about 3 on the volume knob.

I built it using Carbon comp. resistors, getting ready to replace the anode resistors with metal film and see if that gets rid of the static.

I have not put it on a scope yet. I am getting ready to tear it completely down and start over from scratch. You get yours sorted out, I am going to be yelling for help on mine. This is a pretty simple amp (component count) and should be easy to trouble shoot but it has whipped me.

[labb]

Tube depot has the NOS 6SC7's for 19.95 each. shipping is $4.50 by fed ex. Last ones I bought were Westinghouse.

[Cliff Schecht]

Labb it sounds like you went through the same problems that I went through with my 5C3 build. I spent quite a while getting this amp tuned in and getting the noise/oscillation problems cut down significantly. In the end I went with DC heaters on the whole amp (this comes with its own set of problems) and rewired some of the input grounding scheme to really squash noise and kill the oscillations. I didn't end up using any band-aids (capacitor strapped plates and such) but did end up rebuilding parts of this amp a few times before getting it "right".

As I mentioned earlier, one thing that I found necessary was the diode clamping mod (aka Paul Ruby mod I think) at the output stage to prevent the bias from shifting when the output stage is pushed into distortion.

I did not ever end up using any input resistance on the grid of the phase inverter to prevent it from clamping as Merlin suggests, I scoped this part of the amp and never found the PI had any issues when pushed hard.

Because the input is grid leaked, the gain is as high as it can be and the lead dress around here is important. I had to be careful about my grounding scheme and in the end I ended up using isolated input jacks to route the grounds of the first stage right to the "-" side of the e-cap that decouples this stage. This was because I used a chassis ground scheme in a steel chassis, something I will probably never do again unless I plan it out properly. Not only does the chassis carry large currents from the speaker output which can leak everywhere, steel chassis tend to couple stray flux from the power transformer which can get in everywhere as well. Since this chassis was a recycled 50's hifi amp, my layout was predetermined and I tried to work with what I had. My first attempt at this build was quite messy (I rushed it) and I ended up rebuilding the entire input stage to fix a lot of my problems (seems that shielded wire runs didn't really help ever).

[labb]

I have tried two different grounding schemes in this amp. First was buss ground for every thing with the buss grounded right at the input jacks. That is the way I did the first one and it is quiet. Not this one. I went back and separated the power section from the pre amp section. The filter caps, the HT CT, and the power tubes cathode are all grounded on a tab at the first filter cap. The filament ct is connected at the cathode of the power tubes to elevate the heaters. The PI, pre amp, vol/tone pots and input jacks are grounded on a tab at the input jacks. The OT I am using (which is suppose to be a replica of the original) has the negative grounded to the frame of the OT so the speaker is just grounded at the speaker output jack. I did run shielded cable from the vol pot to the grid of V2. I don't get much hum until everything is dimed. I may hook a 6 volt battery up to the heaters and see if it changes anything.

The Paul Ruby mod you reference, is that the same one that is used on the Marshall 18 watt?

I like building amps. I like trouble shooting and repairing amps. But this little amp is starting to get to me.

[tubeswell]

I have tried two different grounding schemes in this amp. First was buss ground for every thing with the buss grounded right at the input jacks. That is the way I did the first one and it is quiet. Not this one. I went back and separated the power section from the pre amp section. The filter caps, the HT CT, and the power tubes cathode are all grounded on a tab at the first filter cap. The filament ct is connected at the cathode of the power tubes to elevate the heaters. The PI, pre amp, vol/tone pots and input jacks are grounded on a tab at the input jacks. The OT I am using (which is suppose to be a replica of the original) has the negative grounded to the frame of the OT so the speaker is just grounded at the speaker output jack. I did run shielded cable from the vol pot to the grid of V2. I don't get much hum until everything is dimed. I may hook a 6 volt battery up to the heaters and see if it changes anything.

The Paul Ruby mod you reference, is that the same one that is used on the Marshall 18 watt?

I like building amps. I like trouble shooting and repairing amps. But this little amp is starting to get to me.

Try a DC supply. The octals don't have hum-balancing filaments like the noval dual triodes do.

[Cliff Schecht]

Don't get discouraged man, mine took me probably 6 months of tuning and tweaking before it was where it's at now. This is a finicky design that is touchy about layout, especially because of the lack of filament noise bucking and that grid leaked input.

With the DC heaters, I lucked out in a few ways. This turns out to be tricky to implement for a few reasons. My old transformer was wound for 117V which means the heater voltage comes out at maybe 6.8V AC which, after rectification and losses was close to the 6.3V DC I was aiming for. If you are using a transformer wound for modern voltages (120-125V) then the AC coming out won't rectify to 6.3V like you would want it to, it'll be in the mid to high 5V range.

Something you have to be careful about when using the same transformer that power the rest of the circuit is that the DC heater circuits can throw noise into the other windings. When you run DC heaters you typically have a very large cap right after the rectifier (like 470-1000uF per volt minimum). This large capacitance is what causes issues. When you hit a quick, hard note, the DC filtering cap dumps its energy and then tries to recharge quickly. As it recharges, it pulls extra current from the transformer as quickly as it can and literally causes pulse-like voltage spikes because you have a quick changing current into an inductor (V=L*di/dt) that develops a proportional voltage. I don't remember how I eliminated this aside from reducing the capacitance to an acceptable level and maybe using multiple stages of filtering as well as some series resistance to slow down the cap charging a bit.

Also worth note with DC heaters is that the entire amp doesn't need to be run on DC. The later push-pull stages inherently cancel hum and even the second stage isn't as susceptible to hum as the first stage because the SNR at this stage is already much better than the first stage. It's still a good idea to run all of the preamp stages of this amp on DC because it's only 600mA max that you're looking at but the output stage is unnecessary. I run DC heaters throughout my amp simply because this is what ended up working for me after a few different attempts (and much reading/learning). I can crack open my amp and see what I did if you want though..

[surfsup]

Cliff i have been researching DC heater supply circuits and read the other day (somewhere 1000 links ago) that the initial resevoir cap should be small with a second cap (part of a RC filter) should be the larger cap. Using the existing winding on a PT that puts out higher filament voltage might notb work anymore due to the extra R in the circuit but thought i'd mention that.

Also, i havent looked into this ruby mod at all but is it only for P-P output tubes? Or can it be applied to SE?

[Andy Le Blanc]

with that inverter its pretty easy to tune in the gain with the plate resistors.

you can also balance the DC at the plates, and balance the AC at the divider that sources the inverting signal.

Makes for a much smoother amp, without wrangling over tube types