Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Won't 4k ohm OT in Lightning eat EL84s?

2/4/2000 5:10 PM
Won't 4k ohm OT in Lightning eat EL84s?
Sorry to return to a thread that was hashed out a week or two ago, but I've been looking more closely into OT choices for my homebrew Matchless Lightning 15 type thing (AC15-ish) -- altered preamps, but basically same power stage to be copied right out of an actually chassis point by point (accurately, I hope!).  
Participants in that thread were quoting Spitfire OT measured at 3800 ohms (call it 4k) but techs and transformer winders here in the UK have been telling me they think that would eat tubes. Any thoughts behind this? Did Mark Sampson take account for this elsewhere in his OT or power stage designs, perhaps?  
I want to get close to that quality of sound and other science posted regarding a 4K ohm primary load seemed to make sense, but everybody over here is telling me, "Well, 8K is standard - and safest."  
By the way, I've found two reputable tranny builders here in the UK who will wind them for me either way for reasonable money: Douglas Electronic, who built the transformers for the Orange AD15 (recently and "editor's pick" for gear of the year in Guitar Player); and Majestic Transformers who can wind a one-off to order from original AC15 specs which they have on file. Both are spec'd as 8k, but will do 4k for little or no extra cost.  
Sorry -- this thing got long-winded, as is usual for me. Many thanks for any help out there.  
PS - wanted to put this in the Guitar Amps forum, but doesn't seem to be taking entries just now. Apologies in advance if multiples suddenly turn up!
2/4/2000 5:37 PM
higher primary z will indeed put more stress on screens, but will end up causing less peak cathode current due to the higher impedance. properly sized screen stoppers should limit screen dissipation without sonic penalty. you can always "mismatch" secondary z in order to reflect a different primary. i would say go for the 8k primary z since that will give you a higher step down ratio, which can come in handy... you will be able to drive lower impedance secondary loads, and you will also have a better damping factor when operating without negative feedback.  
also, all other things equal, the higher primary z unit will probably have higher primary inductance, leading to a lower frequency LF rolloff.  
2/4/2000 9:10 PM

What does he mean he'll wind it at 4k for little or no extra cost? To wind it from 4k up to 8k is about another 25% increase in wire length. How could less wire cost more?  
If you like the sound of the Spitfire or Lightning, then you want a reflected load of 8ohm/4kohms. (I'm the guy who measured the turns ratio at 21.8, which means an impedance ratio of 3800 plate load with an 8ohm speaker load.) The 8k:8 load will not sound the same as a Spitfire or Lightning. It won't sound bad, but a little more well behaved. 8k is the textbook load with p-p EL84s for max power output at min harmonic distortion (high fidelity). These smaller Matchless amps have a unique harmonic content, and the surprisingly low load is a main player in the equation. They do run the tubes hard, but that's the tradeoff for superior tone. EL84s are relatively inexpensive anyway.  
If you use a transformer with an 8k load, and it has a larger core (more iron) than the 4k transformer, the low end cutoff frequency might be as low as the 4k transformer. You don't need response below 100hz on a lead guitar amp anyway, but the iron on that little amp is much larger than a Blues Jr output transformer, even though the primary load on the Matchless is lower than the Fender.  
2/4/2000 9:55 PM

Huh!Sorry Doc,but the fundamental of the low E is 82.4hz,then you do need to go under 100Hz,not much though.  
Best regards.  
2/4/2000 11:02 PM

>>To wind it from 4k up to 8k is about another 25% increase in wire length. How could less wire cost more?<<  
Depending on how the windings lay out, he could just be cutting the secondary winding turns down a bit.. 6 of one..  
FWIW I wouldn't worry about the Fo issue with this at all.. look at freq resp of Marshall 4x12 (king chunk!) on the Aikenamps site. Down a ton by 100Hz. F1 on guitar low E is often perceived (and even notated) up an octave anyhow. I'm not a believer in the whole 100hz thing myself.
2/7/2000 9:43 AM

In fairness to the guy at Douglas transformers, the lower price of the 8k OT was due to the fact that it was coming off a large OEM run they were doing for Orange anyway, so he could do me one of those for 18 (about $28 - not bad for the quality and specs of OT). To do me a 4k OT at those specs they'd have to wind me a one-off, so it's the labor, not the wire, which I'd be paying a little extra for.  
Otherwise, I like the reasoning... and the whole idea is to build an amp that's somewhat different from the textbook designs, with some superlative harmonic content, which seems to me the only reason to try to "copy" a Matchless anyway. And what the hell, I've got plenty of cheap Sovtek EL84s around to burn out anyway! (along with some good Philips and even two or three pairs of Mullards: presumably these good NOS tubes will "take the heat" a little better and longer?).  
PS - Sorry for not responding for two days -- I'm away from my office-bound Web access over the weekend...
2/4/2000 10:57 PM

S/B 8K IMHO. 4K is rough on the Anode diss.. Sampson apparently just liked the sound and tossed the theory out the window. I know guys who swear by 4K loads in 18W amps (using AC30 OT's FWIW) but I can't reccomend the practice.
   Page 1 of 4 Next> Last Page>>