This is not the case , shortly , you need high inductance ( big iron ) for low frequency and low leakage inductance for high frequency ( complex interleaving ) .
So you dont need big iron , but you still need good interleaving . Even with low number of turns you still need interleaving
Last edited:
Depanatoru: Are you saying that the same interleaving is required in order to achieve excellent quality high frequency performance regardless of the what the low frequency requirements are? What about core size?
Absolutly , this are two different problems , you must have interleaving for high frequency . The leakage inductance of the transformer is in series with the anode of the valve , if it's sufficiently high ( like it is with no or poor interleaving ) at 20KHz the inductive reactance will greatly limit the power .
Bigger core and larger winding window enables you to use thicker wire ( in primary and secondary ) in accordance with power output . So you get the same inductance or higher with fewer turns . And the big core also won't saturate that easily of course . So it is both for overall power output and bass .
Bigger core and larger winding window enables you to use thicker wire ( in primary and secondary ) in accordance with power output . So you get the same inductance or higher with fewer turns . And the big core also won't saturate that easily of course . So it is both for overall power output and bass .
Last edited:
Large transformer is not necessary for HF. It is like SMPS, where small transformer can handle large power. Because of smaller size, low primary inductance with small number of turns, and compactness of windings, leakage inductance is low even without interleaving.
BTW, it will be a true transformer, that is a something with true magnetic coupling, not a highly interleaved TLCD (Transformer-Like Coupling Device).
BTW, it will be a true transformer, that is a something with true magnetic coupling, not a highly interleaved TLCD (Transformer-Like Coupling Device).
Of course the transformer for high frequency can be much smaller , but power is power , you can't use the same thin wire for 5W and for 50W . Than you need a bigger core where you can fit thicker wire ...
With fewer turns the leakage inductance is lower , but not enough to be good , just try a power transformer wich is exactly like this , fewer turns and lower inductance .
No transformer without interleaving is good for hi-fi ... maybe some toroidal transformers . But even then you would want interleaving for even better quality ... or just want to do it as cheaply as possible ?
With fewer turns the leakage inductance is lower , but not enough to be good , just try a power transformer wich is exactly like this , fewer turns and lower inductance .
No transformer without interleaving is good for hi-fi ... maybe some toroidal transformers . But even then you would want interleaving for even better quality ... or just want to do it as cheaply as possible ?
Last edited:
This is extremely wrong. Core coupling won't save you from winding geometry, and the later determines HF parasitics, such as Ls, Cs and Cp + resonances.
You might get away without interleaving only if the number of turns gets really small, to fight off leakage inductance.
Power (from the POV of flux density vs frequency) increases in a square fashion vs frequency. For example a 5W at 25Hz OPT will be able to do 1280W at 400Hz.
Last edited:
tizman wants to cross around 32 times higher than 20Hz. If we double inductance to account for the octave of the cross, that's 0.25 the number of turns. Not to mention that this needlessly quadruples the current the core can support, and that the output device needed will probably be small compared to one used for bass.
If you bi-amp, then it really doesn't matter that the treble amp has a full range OPT because there is next to no current drawn for frequencies below 630hz, even if the amp is before the crossover.
The HF OPT would be handling 400 HZ and above. Given this frequency of operation, a primary of 5000 Ohms and a secondary of 8 Ohms, and an output of no more than 4 watts, what would an OPT with these required specs look like? I have an Electraprint 3 watt OPT designed for 45 tubes that is quite large, but it is designed for full frequency range operation. I would think that if it were made for 400 HZ and up, that it would need to be much smaller.
Does this mean that there is no benefit to providing the HF amp with a signal that has had the bass frequencies removed? I was toying with the idea of putting a passive line level crossover before the HF amp, but I won’t bother if there is no benefit.
If we assume you will be using HiB core, where the magnetizing knee starts at nearly 1.4 flux density and you will be using 0.7T for DC magnetizing current (SE OPT).
4W at 5k primary impedance are reached with 142V Rms on the primary. Let's assume you will be using a tiny (2sq.cm) HiB core, then you'll need 570 primary turns.
4W at 5k primary impedance are reached with 142V Rms on the primary. Let's assume you will be using a tiny (2sq.cm) HiB core, then you'll need 570 primary turns.
I think it would be cool to build an amp that is purpose made to drive a two way speaker. An amp with four channels, with 2 pairs of tubes and two pairs of OPTs. KT88 and 6V6? 300B and 45? Maybe a mix, like KT120 wired ultra linear for bass frequencies and 45 tubes for the high frequencies? Or just two separate amps. The bass amp can use any large full range OPT and also be used as a regular amp. The HF amp can use a pair of OPTs optimized for HF.
“4W at 5k primary impedance are reached with 142V Rms on the primary. Let's assume you will be using a tiny (2sq.cm) HiB core, then you'll need 570 primary turns.”
Thanks for info! What does 2 watts look like? I have taken a couple of OPTs apart, but have never built one. A 2 sq. centimetre core seems very small. Is that even big enough for 570 primary turns?
Thanks for info! What does 2 watts look like? I have taken a couple of OPTs apart, but have never built one. A 2 sq. centimetre core seems very small. Is that even big enough for 570 primary turns?
I had this question some years ago, and that resulted in this thread, with some nice information.
Output transformer for 100Hz and up
Post 22 tells some details of the OPT I got. I must admit that I did not do very extensive comparison with fullrange OPTs, yet.
An off the shelf transformer I have been interested in is the LL2766 50mA from Lundahl, only 750gr/piece. My idea is to put the two primaries in parallel, to get a 40:2 OPT with about 12H primary, or about 2400 ohms (gap at 100mA) on a 6 ohm load. Per Lundahl mentioned it could work, but never tested it. I did not either, as I spent the last years caring for my little babies (the actual ones, with legs and arms and a crying head, not the amp babies) and spending the spare time on loudspeakers. But I still like the idea of the lundahls, some switched HV PS would make for a really small and leightweight amplifier.
Another thought I have. Lots of manufacturers have obtained the expertise to make their fullrange transformers through extensive prototype units. Expertise in HF transformers I have never seen, and they surely can do something good on basis of their general knowledge, but they are not going to wind 10 prototypes to optimize the performance, I think.
Output transformer for 100Hz and up
Post 22 tells some details of the OPT I got. I must admit that I did not do very extensive comparison with fullrange OPTs, yet.
An off the shelf transformer I have been interested in is the LL2766 50mA from Lundahl, only 750gr/piece. My idea is to put the two primaries in parallel, to get a 40:2 OPT with about 12H primary, or about 2400 ohms (gap at 100mA) on a 6 ohm load. Per Lundahl mentioned it could work, but never tested it. I did not either, as I spent the last years caring for my little babies (the actual ones, with legs and arms and a crying head, not the amp babies) and spending the spare time on loudspeakers. But I still like the idea of the lundahls, some switched HV PS would make for a really small and leightweight amplifier.
Another thought I have. Lots of manufacturers have obtained the expertise to make their fullrange transformers through extensive prototype units. Expertise in HF transformers I have never seen, and they surely can do something good on basis of their general knowledge, but they are not going to wind 10 prototypes to optimize the performance, I think.
The HF section is a Renkus-Heinz SSD-1800 compression driver in an Altec 511B. I’m guessing it’s around 107 DB/W/M with the HF shaping in the crossover.