I and others have done similar things on other amplifiers, and at worst to get something that looks nice you just order a nice piece (or 2) of copper or stainless steel plate in the appropriate dimensions online, measure and drill for the new transformers and for some of the mounting locations of the old transformers - if done with care you both stiffen the flimsy AQ chassis and end up with something that looks nice as well.
ASL HK makes very reasonable quality OPTs, but the Tamuras are still likely to be a rather significant improvement if you purchase one of their higher end products.
ASL HK makes very reasonable quality OPTs, but the Tamuras are still likely to be a rather significant improvement if you purchase one of their higher end products.
This is serious power game. I presume you are upwards 520V B+ region. You won't get optimum throughput power with this Prim Z. You should be aiming around 2K A-A and you will have to take care of ringing with A and g2 snubbers. The best tranny I've used is the Sowter 8659 43% UL with 4 sep o/p windings for optimum quality and b/w.. Study the Sowter website p-p secondary configuration.tigro said:
Beware they aren't cheap !
rich
Tigro;
The 8659 or variants are large transformers (?10Kg )and particulary smooth response using fixed bias around 60mA per tube. At 2/3 power I got a 10Khz square wave looking in pic. It all depends on the performance of the driver stage. When you get the power stage performing the square wave shown in pic, then you can't ask for more. The distortion is also very low.
richj
The 8659 or variants are large transformers (?10Kg )and particulary smooth response using fixed bias around 60mA per tube. At 2/3 power I got a 10Khz square wave looking in pic. It all depends on the performance of the driver stage. When you get the power stage performing the square wave shown in pic, then you can't ask for more. The distortion is also very low.
richj
Attachments
Bandersnatch said:
Both power amps in pic are my strong man studio "Samson & Goliath" class both capable of cruising 200W per channel with KT88's, one has a Sowter 8659 14 sect and other Majestic Partridge spec'd 18 sect, both British wound o/p trannies performing -3dB down around 50KHz. On listening both are close, -, --> I favour the Majestic on iron/winding technicals; requires less A-G2 snubbing but is 10% larger, more consistent construction and goes down to 15Hz and this one has reduced B+ running Tung-Sol 6550's (Rt side).
These heavy amps (each 55kg) are for those who like their music blinding loud or play clean MI (as I do): they use a variable B+ with a PFC converter for economic running. The 12BY7A driver stages & front end Concertina ECF80 are identical.
Those wanting a real UL p-p design challenge in the high end game have got one..every aspect must be math'ed out, including loop gains. Both amps can accept 35dB global nfb before instability sets in. This is a tough challenge as the winding parasitics are much higher than smaller designs, require careful snubbing as the energy is far higher and these amp designs cannot be slapped together as with lower powers. A simple mistake can be expensive in both tubes and patience.
Take note!
richj
Attachments
Tigro; enclosed are some early loop gain graphs: The upper Majestic is far more consistent. Both examples sampled over modest powers.
Note, as the transformer size goes up, so do the parasitics which effect the upper bandwidth far more than with lower power examples and relies on nfb to correct response. As mentioned regarding snubbers; a 10KHz square wave pushed through at 100W ++ will easily destroy 3W rated stuff, so components must be rated accordingly. Hint: The driver stage must not slew before the output stage and class A balanced drivers will sink min 30mA per pair.
Hope this helps
richj
Note, as the transformer size goes up, so do the parasitics which effect the upper bandwidth far more than with lower power examples and relies on nfb to correct response. As mentioned regarding snubbers; a 10KHz square wave pushed through at 100W ++ will easily destroy 3W rated stuff, so components must be rated accordingly. Hint: The driver stage must not slew before the output stage and class A balanced drivers will sink min 30mA per pair.
Hope this helps
richj
Attachments
Dear Rich
Thank you for your detailed and well documented answer.
I use an existing design, it's an Antique Sound Lab AQ 1008DT
monoblock and I want to upgrade the output transformer.
I will use Sowter U059 model and have it modified for 2.25K
primary impedance insrtead of 1.75K.
Regards
Tigro
Thank you for your detailed and well documented answer.
I use an existing design, it's an Antique Sound Lab AQ 1008DT
monoblock and I want to upgrade the output transformer.
I will use Sowter U059 model and have it modified for 2.25K
primary impedance insrtead of 1.75K.
Regards
Tigro
Here is the Antique Sound Lab site, I no longer see any transformers listed there since the redesign so you will need to email them and find out if they still sell a line of transformers. (As I've indicated they used to make some good ones, they wind all of the transformers and chokes used in their products and used to offer a selection at retail as well.)
http://www.tubehifi.com/main_page.htm
The odd choice of text and background colors make their site a very difficult read on my computer.
http://www.tubehifi.com/main_page.htm
The odd choice of text and background colors make their site a very difficult read on my computer.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.