....and Now for Something Completely Different!

[David Root]

"Could one not assume that IF your PT delivers 250 ma that the 100 watt OT would work in ghetto mode / parallel?"

John, if I understand you correctly, that's what I'm assuming. If OTOH 250mA is not enough I can use a bigger PT. I have a BF Twin PT, for example, but that is surely overkill, no?

[John_P_WI]

Hi David, yes we are in agreement.

[David Root]

Back to the amp I just debugged....I finished cleaning up my shop, it had got so bad I couldn't find stuff any more.

So I changed the KT66 screen resistor as someone commented that it's too big for UL operation. Was 5K1 3W metal oxide. Is now 360 ohm 2W carbon comp.

Sounds MUCH better! More open, bigger, more "harmonic", yet still crunches nicely when pushed. I've been told carbon comp makes the best screen resistor and I'm now a believer.

[David Root]

Hi Matt, very interesting stuff, as usual.

My 125E is currently rated by Hammond at 80mA, and I'm drawing 85mA quiescent, referenced to 120VAC, 25.4V cathode, 300R. My wall AC is usually nearer 122V most of the time, so I'd be at 86mA. Plus a little bit more cranked, maybe 90mA?
So yes, Hammond underrates their iron as I noted no signs of stress.

Going to my 2xKT88 "fun on paper" (at least so far), the PT I spec'd is an Edcor, not a Hammond. I've used Edcor iron before and I think it is every bit the equal of Hammond and probably a bit better. The voltage rated is under full rated load. I came up with the PS voltages from PSUD2 using a 7.5% regulation factor.

I haven't bought that PT yet so I can always go to a 300mA PT to maintain the 350-ish volts plate that sounds so good with the KT66. Who knows, maybe 400V would be better with a KT88?

On the OT, I'm having a hard time believing that a 7 pound Marshall JMP type OT designed for four cranked EL34s can't reliably handle 250mA through it in this circuit. Yes I know it has 350V DC on the primary and it's somewhat unbalanced too, but Mag Cpts makes an old school OT Made-in-Chicago, that should count for something, no?

After all, Marshall put 4x6550 on those '70s OTs for the US market, and I haven't heard any horror stories about them not being able to handle four cranked 6550s.

Coming back to the KT88, let's say they draw 125mA each. At 350V that's close to 44W plate + screen, which is indeed pushing it--for the tubes. Maybe my pair of JJs won't last too long?

[David Root]

Thanx again matt. I feel a bit more at ease with this now. Best thing I can do at this point is put it in my build plan and check I have all the major parts.

The 42+8W Pd does bother me a bit because that is for an MOV/GEC tube and I'll be using (at least starting with) a pair of ten year old (but unused) JJs. Based on what I noted y'day on screen voltages on the single KT66 after I changed the screen resistor to 360 ohms, maybe 0.5V across the resistor, it might be drawing 1.5mA, so essentially all the dissipation is in the plate. It sounds like this might increase a bit in the KT88 but with such a small screen resistor I can't see screen draw being much of a component of total Pd.

The 20K plate resistor on V3 I came up with by plotting on the plate curve. These TV tubes are not 5879s after all, and draw more current, but I'll take another look at that and the bootstrapping CF too.

I'll also see what I can find out about Edcor OTs in this application too. They do actually make SE airgapped OTs with a 40% UL tap, just in case! I'll also take another look at Hammond's line of PP OTs too but IIRC their primaries are too high.

[David Root]

Theoretically the 42 and 8 is common to all KT88s; what I was getting at is modern construction KT88s may not in practice be that tough, so it might be a good idea to back off the Pd a little to avoid a fireworks display.

Why KT88s? Because they're in stock, along with the Marshall OT, and I wanted to see what a 30-40W version of the KT66 amp would sound like, without spending a grundle of $$$.

Edcor does make UL SE OTs with enough oomph I think. They have one, model CXSE25-8-1.25K, primary 1.25K, explicitly rated at 250mA, also rated 25W 20-20kHz -1dB, which should handle 35 to 40W of guitar output OK. Weighs almost 10 lb. $88. Comes in 4 or 8 ohm sec. 40% UL tap.

[David Root]

Finally-got my friend Chris Rolin to do some soundclips of the Ghetto UL SE KT66 amp. He liked it so much he's putting it on one track of his CD that he's recording now. He used my Dumble -124 on some others.

I made notes on settings etc if anyone's interested. Basic equipment is the amp into a single G12-80, SM-57 mic into an LA-610 Mk II. Slight compression (3) on the LA-610.

[Colossal]

Great job David!

[10thTx]

Man, I am VERY impressed with how good this amp sounds!!!

Congrats! You've come up with a true winner, IMO.

Nice tone. And it sounds like it is very touch sensitive to play?

With respect, 10thtx

[EtherealWidow]

^^ I was thinking the same thing. The thing has great attack.

[David Root]

Thank you folks and thanx to Geezer for the original inspiration, the first half of the amp is a straight HoSo56 w/12AU7 CF. And of course the Ghetto UL! I just put them together with a KT66 and a 5879 driver.

It is very touch sensitive at the settings used. All the clips except one were recorded with the gain on 2:30PM and MV (Trim) on 9:00AM. The Benedetto track was gain 11:45 AM, trim 09:30 AM. Bass on 8, mids/treble 4 1/2.

I probably need to put a bigger cathode bypass cap on the KT66 so the bass can be leveled a bit better.

Guitar on all except Benedetto was an '89 Warmoth parts thinline Tele w/ Bill Lawrence pickups. The Benedetto guitar is the Bravo model archtop.

[Structo]

Wow, sounds great David.

Also doesn't hurt that you are a pretty darn good guitarist as well.

[David Root]

Thanx for some new perspective matt. I'm not looking for "big bass" so much as a little bass boost at the end of the chain so the bass pot doesn't have to be on 8 to get the warmer tone that Chris Rolin (the player) likes. I am thinking of trying a 50uF 100V bypass cap on the KT66 cathode.

Yes the TS favors the mids with that 100K slope resistor. That is about the only Dumbley thing in the amp! I could change it to 51K, that would almost equalize the bass & mids. But I wonder if that might alter something in the driver, whereas the bypass cap on the KT66 won't.

And yes, the amp can do much more overdrive than that. You get a hint in that direction on the Rock Chords track, sounds more Marshally I think. But it has turned out to be very versatile and great sounding, so I'll be happy to take credit for the 5879 driver as that was all mine!

Tom, I'm not the player unfortunately, I am not much of a player sad to say. My friend Chris Rolin is a pro musician and music teacher, and has a nice home recording system to boot. He owns my #124 Dumble build.

[Meat&Beer]

...That jazz track!

Simply gorgeous, very well done, sir.

[David Root]

Thanx again folks! If anyone wants to try it, here's the current 10-25-13 schematic & layout.
The layout is easily adapted to eyelets rather than turrets. I normally use eyelets but I had this long turretboard from a previous build that didn't cut it, so I just bandsawed it into two halves for this build.