• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Output transformer for 100Hz and up

Status
Not open for further replies.
Jump to Latest
E

ErikdeBest

Member
Joined 2003
#1
This has been discussed here before, but I would like to bring it up again.
I have a pair of active loudspeakers, where a Visaton B200 plays from 400Hz upwards. Crossover is active, and the amplifier uses "Fullrange" output transformers.

These transformers are monsters, with enough inductance to play at 20Hz and less. As I don't need that low end, and based on the bit of transformer technology that I know, I am thinking it would be interesting to employ a transformer with a -3dB point at say 100Hz. The advantages I see is that they would be lighter and have lower insertion loss (as they need less wire).

I know guitar amps use transformers with a -3dB at 80Hz or so, but I am thinking about something more "high end", that could outperform the transformers I am using now (amorphous double C core from AE Europe).

Does someone have experience with such transformers – and a recommendation that would be able to wind a pair of them for me?

Many thanks and best regards,
Erik
 
#5
Erik,

Look for full range "iron", with a superior top end, of the proper primary impedance, but of reduced power handling capability compared to the OEM parts.

Physically smaller "iron" can handle the amp's power, as long as the demand is not in the deep bass.
 
#6
i think any custom winder would understand this approach and calculate / construct them accordingly. also nickel and amorphous cores could shine here, since steel is still the domain for a bass. you are right about the smaller size, it will positively affect mids/highs. and/or you can tolerate bigger gap for your current, if you choose existing transformer.
 
#9
I think the transformer will emit distortion if the signal saturate the core with low frequency, this is not good for the feedback of the amplifier and I would not like to hear the rest of what the output is... Very nice impractical idea.

Do filter the bass at the beginning of the amplifier if you want to keep your idea.
 
#11
I'm going to assume this is probably an SE amplifier, that feedback is not employed and that there is either an active or passive XO between the source and amplifier..

I am working on a similar amplifier design for a friend and plan to ask my supplier to design and manufacture some output transformers that are -3dB at 100Hz which will allow a reduction in core size (double c core) and in theory at least an improvement in HF performance and bandwidth. They will certainly be a lot lighter, more compact and exhibit lower copper loss, and parasitics.. The amps in question will operate from about 400Hz upwards and will see little below this frequency.

Perhaps AE would be willing to wind you restricted bandwidth amorphous core outputs for this application. I would also check with Monolith Magnetics in Belgium.
 
#12
gabdx said:
I think the transformer will emit distortion if the signal saturate the core with low frequency, this is not good for the feedback of the amplifier and I would not like to hear the rest of what the output is... Very nice impractical idea.

Do filter the bass at the beginning of the amplifier if you want to keep your idea.
Click to expand...

thanks for the reply. As I wrote in my first post, I will be using an active X-over before the amplifier, so it won't see any LF signal.
 
#13
kevinkr said:
I'm going to assume this is probably an SE amplifier, that feedback is not employed and that there is either an active or passive XO between the source and amplifier..

I am working on a similar amplifier design for a friend and plan to ask my supplier to design and manufacture some output transformers that are -3dB at 100Hz which will allow a reduction in core size (double c core) and in theory at least an improvement in HF performance and bandwidth. They will certainly be a lot lighter, more compact and exhibit lower copper loss, and parasitics.. The amps in question will operate from about 400Hz upwards and will see little below this frequency.

Perhaps AE would be willing to wind you restricted bandwidth amorphous core outputs for this application. I would also check with Monolith Magnetics in Belgium.
Click to expand...

Hi Kevin,

thanks for your reply! Your assumptions are right, I am indeed using an active crossover and only limited local feedback.
I will contact some of these winders and ask them about previous experience with restricted bandwith transformers.

Erik
 
#16
Hi Seppoa,
I know the Indel transformers, and have used some of them with good results - but the OPT from AE outperform them (but hey, the AEs cost way more).

Still the Indels are also designed with lower frequencies in mind, and I am really willing to sacrifice this to obtain way lower amount of copper and core material for handling a certain amount of power.

cheers, Erik
 
#17
Hi Erik,

Any Push Pull OPT has a constant wich is the square root of the power divided by the frequency.
For exemple an OPT designed for 10W @ 20Hz is able to transfer 40W @ 40Hz with the same level of distortion because in both cases the induction is the same.

For an SE OPT you must know the fraction of induction that the designer has "reserved" for DC premagnetisation, usually one half of the total.

Copper losses may increase a bit but this just produces heat, no distortion 😉

Yves.
 
#18
IMHO you need a transformer with -3Db point at least one octave lower than your actual amplifier range.
Otherwise you will have too many phase shift.
Also you can build a high pass filter in the amp so that the transformer core will work at lower induction in a more linear zone.
This also avoid the use of an external electronic X-over. The simpler, the better 😉

a OPT with say, 50 Hz - 100KHz bandwidth is so much easier to design than one with 10Hz - 30 KHz.

In my experience a skilled designer could do an excellent work without the extra cost of an amorphous core (best quality FeSi works better for SE and costs a little bit less)

I would consider Permalloy if the DC current is not too high.
 
#19
Yvesm said:
Hi Erik,

Any Push Pull OPT has a constant wich is the square root of the power divided by the frequency.
For exemple an OPT designed for 10W @ 20Hz is able to transfer 40W @ 40Hz with the same level of distortion because in both cases the induction is the same.

For an SE OPT you must know the fraction of induction that the designer has "reserved" for DC premagnetisation, usually one half of the total.

Copper losses may increase a bit but this just produces heat, no distortion 😉

Yves.
Click to expand...

Hi Yves,
thanks for your reply. Indeed, I remember this math from a previous discussion about transfomers here at Diyaudio.
I am aiming for 10W at 200Hz - that could be a really small transformer, I think?
Erik
 
#20
CiroMarzio said:
IMHO you need a transformer with -3Db point at least one octave lower than your actual amplifier range.
Otherwise you will have too many phase shift.
Also you can build a high pass filter in the amp so that the transformer core will work at lower induction in a more linear zone.
This also avoid the use of an external electronic X-over. The simpler, the better 😉

a OPT with say, 50 Hz - 100KHz bandwidth is so much easier to design than one with 10Hz - 30 KHz.

In my experience a skilled designer could do an excellent work without the extra cost of an amorphous core (best quality FeSi works better for SE and costs a little bit less)

I would consider Permalloy if the DC current is not too high.
Click to expand...

Hi Ciro.
The B200 plays from 400Hz upwards, therefore I "specified" the -3dB point at 100Hz, two octaves lower. I indeed want to include the X-over in the own amplifier, playing with the RC's between stages - but at the moment I am still optimizing some settings 🙂
Cheers,
Erik
 
Status
Not open for further replies.
Top Bottom