cause of you , i have S9 summit MonolthMagnetics on the way 🙂
in a couple of week , i will see how they will compete with the lundhals..
( ps: my pair of LL1664AM are on the swap meet)
Tamura F-2007
IMHO the best SE 2A3/300B TU is TAMURA F-2007. Please read my SE application: Madama Butterfly....
IMHO the best SE 2A3/300B TU is TAMURA F-2007. Please read my SE application: Madama Butterfly....
I'm tempted by your OPTs! But I need good bass, since I have 6" Alpair 10 full-range units in my speakers and they are rather minimal on bass anyway.
Looking forward very much to hearing your views on the S-9. You are doing good work for all of us reviewing these OPTs.
I'm tempted by the ETU-1, simply because it's cheaper. Were you not also tempted by that one? Same core, but Mylar rather than PTFE dialectric.
http://www.monolithmagnetics.com/im...gle ended output transformer prelim rev01.pdf
Last edited:
the lundalh is a really good ,for is performance the price is low ..ll1623 is a classic ,good sound and configurability
Number of "C" pieces alone doesn't say much, what really matters are several interconnected factors - core cross-section, number of primary turns, max flux density (Bmax) and minimum inductance (in H) at Fmin (e.g. 20 Hz).
+ parasitic parameters - leakage inductance, stray capacitance and resonant frequency as a factor of primary turns and winding geometry.
Most manufacturers use 4 "C" pieces and single coil design (~easier to assemble larger core cross-section).
Lundahl - 2 "C" pieces and twin coil symmetric configuration (~easier to get lower parasitic parameters).
None is better or worse, its just 2 different configurations aimed for the same purpose. Finished item performance depends upon particular implementation, engineering skills and build quality.
Hope this helps.
I'm not convinced. Windings in two C core design will always be immersed in *less* magnetic flux when compared to windings in four C core design. And that is that. And if that is not that, then why Tango, Tamura, Hashimoto etc. go into 'trouble' of giving us four C cores per transformer?
Interesting - I didn't know they had 4 cores. The Monoliths have 2. I just checked. Doesn't seem to do them much harm, but they are big transformers, weighing around 5 kilos.
"The heart of any Monolith transformer is a high grade SiFe or Metlas amorphous dual C-core made to our specifications"
"The heart of any Monolith transformer is a high grade SiFe or Metlas amorphous dual C-core made to our specifications"
I hope you like them, I'm very pleased with the MM ITs I am currently running, and am using some other transformers they made for me in other projects. The MM ITs are 4 core.
When it says c-core transformer it's meant to have two core pieces. Dual c-core means four core pieces.
Last edited:
Please explain what is the difference between these 2 designs in terms of "immersion in magnetic flux":
1) 4x"C", 4cm x 4cm = 16 cm2 cross-section (4cm ribbon width x 2cm thickness each piece), flux density 1.4T@20Hz/50W;
2) 2x"C", 5cm x 3.2 cm = 16 cm2 cross-section (5cm ribbon width x 3.2cm thickness each piece), flux density 1.4T@20Hz/50W;
Primary turns same for both.
And why there are cars or whatever commodity of different shapes and colors?
AFAIK japanese are EI.
Pi winding on sigle bobbing is very good, but due complexity $$$ nobody does it
Your example is both misleading and irrelevant. You can input any combination of anything, it's still:
four cores
core-- windings--core--core--windings--core
vs.
two cores
windings--core--windings--core
Or could it be the bean-counting logic that is at work here?
Last edited:
Its your assumption is misleading and irrelevant, in real engineering only real terms and numbers counts, like effective cross-section area, mean magnetic path length, coil length, etc., not some "immersion in magnetic flux".
4 x "C" single bobbin and 2 x "C" (with larger C pieces) twin bobbin variants can be made with almost equal characteristics if effective cross-section area is equal.
BTW, have you ever disassembled Japaneese output transformer?
I did, and as hpeter said, they were EI.
There is a reason for that - EI allows you to assemble stack of virtually any thickness straight out of raw EI pieces, and once effective cross-section area is increased, primary number of turns (as well as number of interleaved sections) are decreased, and therefore, amount of labor.
An example of extremely thick EI stack - output transformers made by Bob Carver for his parallel 300W push-pull amplifiers.
C cores usually sold in default dimensions (e.g. DIN 41309), their size is optimized for average 50/400 Hz usage, often cores with suitable cross-section area have too small window size to accommodate very large amount of winding typical for audio output transformers. Sometimes it leads to awkward solutions like AudioNote tripple C core models.
Last edited:
😱 can explain better ?
They nearly doubled core cross-section area by means of attaching 2 x C pieces (with double ribbon width and thickness that fits into tube) on the side.
Same could be achieved with 4 loops (8 x C pieces), or custom-built C pieces, but AN probably couldn't source suitable plastic bobbin or overpay for non-standard cores.
This is striking example how boutique, top priced manufacturers save on production cost, and opt for unbelievable awkwardness to extort few pennies of profit.