firefly Filament pins 4,5 or 4,5 and 9 ?

[John Lyons]

I built Doug Hammonds firefly and have been trying to get it "right" for a week or so now. The amp sounds great with the exception of  too much hum. Sounds like 60 cycle...

Anyway, after trying a few star ground techniques and not getting any less hum I happenned to look over the AX84 P1 schematic and noticed that pin 4 and 5 get one leg of the Filament voltage and the other leg goes to pin 9 ("filament center tap"). I tried this and Zing!! no hum (or at least ot much at all).

I noticed when I first built the amp that the filaments were not gowing as bright as they should...Now they are bright and the amp is louder now as far as I can tell.

Does anyone know what the difference in using pin 4,5 and pin 4,5 and 9 is?
The amp seems brighter and crunchier now as well. Maybe I can use the conjunctive filter here? Or is it only for single ended amps.

Thanks folks!

John

[Peter Snowberg]

The 12A_7 series tubes have 12.6 volt filaments with a center tap if you want to run them from 6.3 volts. 6.3 was a very common voltage for other tubes so the provision was added to use that voltage.

If you're running with a 12.6 volt filament supply, you want to run the power to pins 4 and 5 and leave pin 9 unconnected. If you're using 6.3 volts, you'll want to bridge pins 4 & 5 and apply power between the 4+5 combo and pin 9. DO NOT apply 12 volts across 4+5 and 9 as you'll burn out the tube very quickly.

If you still have hum, I would suggest a couple of things....

* If you have no connections between your filament supply and the rest of the amp other than a single grounding point, try removing that ground point and connecting it to the cathodes of the 12AU7. That will raise the potential on the filament a few volts which will reduce hum.

* Play with different placements of the transformers because you could be getting hum induced by magnetic coupling between the power transformer and output transformer or output tube.  

* If you have a scope and know how to read signals on high voltage lines with it, look for hum in the B+ supply. You might want to add another stage of RC or LC filter to the B+ supply.

* Try making a DC supply for the filaments using a bridge rectifier, a BIG filter cap, and a power resistor to drop the voltage to the correct level. Adding the rectifier and cap will multiply the input AC voltage by 1.4 so you need to back it off to make a suitable DC filament supply.

Good luck!

Take care,
-Peter

[John Lyons]

Peter.

Once again you bring the info!!!!! So basically I was only getting mimimal (6v  half) the volts on a connection meant for 12.6. It did work but was a bit "browner" and sounded nice, but got the hum. Trade off?

Will the "conjunctive filter"  work with a self-split amp? Or something similar. Ideas to smooth out the harshness?



Thanks.

[Peter Snowberg]

No problem. My pleasure.

So if I get this correctly, you had 6.3V across pins 4 and 5? That would sure make a difference.  

You might have been getting large hum because of the way the AC was interacting with the amp stages. If you have the filament supply grounded via a pair of resistors or the center tap of the filament transformer and then that supply is fed to both heaters.... one heater would be getting the + side of the wave while the other would be getting the - side. The triodes are out of phase already so now adding any noise that is 180 degrees out of phase in successive stages is just going to exacerbate any hum. If you power the filaments in parallel instead... each filament sees the + and the - side of the AC wave at the same time and the hum self-cancels within each triode. That's my theory of what's happening anyway.

If you have harshness you want to cut out, I would advise substitution of two parts to find your desired tone. These parts are speaker(s) and the 12AX7. There is such a huge variety of tones to be had by these changes that I can't recommend adding additional filtering over changing parts. Adding a filter is treating a symptom rather than the problem. If your 12AX7 has small plate structures, I would swap it for one with larger plates (try a Sovtek 12AX7LP or LPS). After that I would try a larger speaker.

If you want to try adding filters, I would start by adding small caps across the 12AX7 plate resistors, either without or with resistors to temper them. After that, I would start with the Zobels (conjunctive filters).

Take care,
-Peter

[petemoore]

Possibly shielding the iron would isolate some of the hum?

[John Lyons]

Aftter playing through the amp this morning I find that I like it as is for the most part. I had been playing it with the wrong filament wiring for a couple weeks and was used to that sound. When I corrected the "problem" the difference was of course night and day! the high end is more forward and the output is louder.  

Funny that in the normal channel the miswire sounded ok, no hum and maybe just a bit dull. When the cascode tube was switched in the hum was there even at low settings. The extra gain just amplified the hum a lot more I guess.

The amp has a .0022 cap across the plates of the output (12au7) anyway... Partial Zobel? I might try another value to use as a switchable filter to get a warmer sound when desired. There is no tone control on the amp...

JOhn

[Doug H]

Aftter playing through the amp this morning I find that I like it as is for the most part. I had been playing it with the wrong filament wiring for a couple weeks and was used to that sound. When I corrected the "problem" the difference was of course night and day! the high end is more forward and the output is louder.  

Funny that in the normal channel the miswire sounded ok, no hum and maybe just a bit dull. When the cascode tube was switched in the hum was there even at low settings. The extra gain just amplified the hum a lot more I guess.

The amp has a .0022 cap across the plates of the output (12au7) anyway... Partial Zobel? I might try another value to use as a switchable filter to get a warmer sound when desired. There is no tone control on the amp...

JOhn

The .002u on the 12au7 plates is a conjunctive filter. You can certainly experiment with it to roll off more highs if you need to, to match the speaker you are using. I also added a 220pf across the plate of the upper cascode triode to roll off the high freq a hair there. It's in my latest schem at my site if you haven't seen it.

Doug

[Paul Marossy]

"After playing through the amp this morning I find that I like it as is for the most part. I had been playing it with the wrong filament wiring for a couple weeks and was used to that sound. When I corrected the "problem" the difference was of course night and day! the high end is more forward and the output is louder.

Funny that in the normal channel the miswire sounded ok, no hum and maybe just a bit dull. When the cascode tube was switched in the hum was there even at low settings. The extra gain just amplified the hum a lot more I guess."

Gee, I guess I'm going to have to fix mine now.  :oops:
I didn't realize that I did the filament wiring wrong until today. It sounds OK at 12.6 volts, except for the hum on the boost channel, as noted. I thought that because it was a Class A push-pull, that the circuit was just more prone to hum... I should have known that I screwed it up. I'm just not fully functional yet with that sorta newborn baby boy around and waking up several times night, still.

Well, at least what's on my site is gonna be straightened out in the next few days. I guess I should do a soundclip of what it sounds like with the 12.6V heater wiring, just for the record.

Sorry for any inconveniences this may have caused.

[John Lyons]

Yes. it does sound ok with the filament miswire. Spongy and a bit smooth...a tad dull in the clean "channel", but nothing objectionable. Just the Hum in th lead channel especially when cranked.

With the correct filament wiring it sound much more forward in the high end and midrange. Bolder too,if you will.

John

[Paul Marossy]

Yes, I'll agree to that. This little amp is one of the few amps that I can stand to play without any reverb or touch of delay.