SR45 amplifier

[johnsonad]

Xavier, I'm glad you hear what we do!  :^)  Made some progress the last day or two in the new design. I'm ready to prototype the power supply and see if my idea is actually workable...

 

[Jim R.]

Xavier,

Congrats on getting the first one done.  I agree they are really special amps.  I'm going to get mine back on a 24/7 schedule today after a few days of being out of commission (me, not the amps.)

I have to say I've never had so much fun with DIY audio, and the rebuilds are going to be even more fun.

Great stuff!

-- Jim

[Grainger49]

Thanks, Xavier!

[Paul Joppa]

Quick update. A couple weeks ago we were talking about the SR45, and Doc B said "if you're going to do a shunt reg'd amp, I want to see it done right and d@%# the cost!" - or words to that general effect. So, after my first enthusiasm led to a $12,000 kit :^) I've been working on something more reasonable. I have a pretty good idea now of where I'm going, and a fairly detailed general layout. I have always wanted to regulate the filament and heater as well as the high voltage, and doing that always leads to a lot of heat under the chassis, which is already the biggest problem with the SR concept. So I'm now sketching quite a large amp with a better thought-out cooling flow and lots of heat sinks. I'm aiming for really long-lasting parts as well, and delay relays for startup and shutdown sequencing.

[xcortes]

I like the approach!

Do I assume 12 x 12 layouts? With three pairs that would mean some serious footprint!

I would like to start sourcing iron. I have the 404s already. Will the bigger chassis allow for, say, Dowdy chokes, or good old BCP15s?

And finally, I guess it's cap coupled? 'Cause then grid chokes would be on the list as well.

It looks like a new car will need to wait a couple o' years!

[johnsonad]

Thanks for the update Paul! Has the project changed from a partial kit to a full kit?

[Paul Joppa]

At this point, I am not allowing for non-standard iron. The chassis is quite large (I'm leaning to 10x16...) but more than half is power supply - there is a little room but not much. Custom chassis plates might be an option. There are several other related developments in the works but I won't discuss them until we have a consensus :^) 

Yes it's cap coupled. Much as I like direct coupling, it adds too much voltage to the power supply - the shunt reg already adds some. Really good caps are available these days!

[Paul Birkeland]

At this point, I am not allowing for non-standard iron.

I think that's a great choice, it would be impossible to design a kit that will work with four different output transformers and five different chokes...

[johnsonad]

Paul, is the 10x16" layout a stereo mono block design?

[xcortes]

"it would be impossible to design a kit that will work with four different output transformers and five different chokes..." No need to do that. As long as my three pairs of TL404s work I'm fine.

Seriously, you know I'll get three pair of these. Using my 404s would just be awesome. But I understand that optimizing the amp for a certain output impedance requires important changes to the ps. So I can only pray to God above that you are designing for 5k!

[johnsonad]

Hmmmmmm, I have a problem with this also that goes a couple of ways. I above all respect and admire what you all do as designers and as a company and the quality of your products has redesigned my system and given me considerable pleasure.

The decision as was posted earlier to use a 4k opt is a little hard to swallow if we are unable to change the operating point. This amp was never meant to be a main stream item nor will most people be able to live with 2 watts output. For those of us that can live with it and want to invest a considerable amount into two watts the choice of 4k leaves us zero options for other iron unless we have it custom wound. Mikey already said he wasn't doing it and to change it to a 5k. There are numerous options in 3k and 5k.  Will you guys please reconsider the end user and change this to 3k or 5k? 

[Paul Birkeland]

Like I said before, I think this design will be perfectly optimized for the iron we use, as soon as you go to something else, you run the risk of the design not being ideal for the iron.

No compromises means no compromises.  All this discussion of which MQ iron you want to go into these amps makes the assumption that our iron will be inferior, which may not be the case in this instance.

[johnsonad]

Paul that's not what I'm saying. I'll use myself as an example. I've exchanged emails with you and PJ both about using the SR45 as a platform for a Stax headphone amp. PJ helped me find the right iron and Sowter modified a preexisting 5k design to make it work. In the 5k range along there is off the shelf and custom options from 2 Ohm to 500 Ohm output in every lamination under the sun.  With the 4k design there is one, yours and where it may be outstanding iron there is no room to be a Bottlehead and experiment.

I'm beginning to feel that you all don't want us to experiment with this amp and that defeats the entire reason for my initial purchase.  If you don't offer us the 5k option, I am out of luck and will need to find a pair of the older SR45 amp that can make work for my particular need or scape it all together.

This being said I will still probably purchase one pair of the new design.

Does Bottlehead intend to release iron in different lams at some point? Are you encouraging winders to adopt your 4k spec?

[xcortes]

OK. I guess I'll wait and see what you come up with. If you use no compromise iron I guess I can sell my three pairs of TL404s (and I'm pretty sure they'll fly). As long as the iron has 4/8/16 ohm taps and I'm allowed to play with the parafeed cap sizes (my crossovers) I'm fine. Otherwise I'll have step aside.

[Paul Joppa]

I would expect the amp to work extremely well with any load impedance between 3K and 5.5K.

I have tried to let the Forum in on what I am exploring as possibilities; I apologize if I have given the erroneous impression that these are commitments or actual product descriptions. To be clear:

***** there is no final product design yet *****

and everything I have said so far is subject to change. My post yesterday is an example - the design has taken a major jump in perspective. This has been creeping up on me for some time, and is the primary reason progress has been so slow the last year or so. My original intent was to modify the original to resolve the excess heat and reliability problems, optimize the operating point, and add DC filament power since an amp of this caliber should not hum, even on extremely efficient speakers. Every solution I came up with has had problems of its own, mostly because my thinking was self-constrained to just modifying the original, flawed (in my mind) design. Once I started to come at it from the idea of "how would I do this to get it right?" it became clear that a significantly different design would have to be the answer.