I'm doing some experiments with standars toroid transformers for my full range ESL project.
The different transformers I have tried are:
Using an old LM3875 amp, 1 Ohm in series with the primary - 9 Ohm paralell(or none) and driving my ES bass panel 250x1500mm d/s 4mm C=600pF.
I get distorsion when I crank the volume up and the transformers start to sing - core saturation ?
Has any one any sugestions/tips on how to best make a DIY transformer capable off 20Hz ?
I'll be crossing over to my mid/high-panel at abt. 2-300Hz, so it "only" has to handle 20-200Hz!
Roger
The different transformers I have tried are:
- Core Windings Ratio
- 4x60VA 2x12/240V 1:80 (in theory)
- 2x80VA 6V/240V 1:80
- 2x200VA 12/240V 1:40
- 2x400VA 5V/240V 1:100
Using an old LM3875 amp, 1 Ohm in series with the primary - 9 Ohm paralell(or none) and driving my ES bass panel 250x1500mm d/s 4mm C=600pF.
I get distorsion when I crank the volume up and the transformers start to sing - core saturation ?
Has any one any sugestions/tips on how to best make a DIY transformer capable off 20Hz ?
I'll be crossing over to my mid/high-panel at abt. 2-300Hz, so it "only" has to handle 20-200Hz!
Roger
240V toroidal power transformers work great for driving ESL at frequencies > 300Hz.
They do not work well for use as bass transformers unless you use many of them and hook the driven low voltage windings in series or choose transformers with higher secondary voltage windings.
Here is why.
Power transformers are designed with the core operating just below saturation(Bmax ~ 10,000).
This is done for reasons of economy, not using more iron or copper than necessary to get the job done.
Bmax = E x 10^8 / (4.44 x f x N x A)
Bmax = Peak AC flux density in core (Gauss)
f = frequency
N = number of primary turns
A = core area cm^2
Note that Bmax is inversely proportional to the number of primary turns.
Typical Bmax for modern transformer iron is about 12,000 gauss at the onset of saturation
So, no matter what VA rating, the number of turns in the 240VAC winding is chosen such that Bmax ~ 10,000 for 50Hz – 60Hz.
What this means is that if you have a 6V/240V transformer and you drive the 6V winding with a 60Hz audio tone,
the transformer core will start saturating at about 20% (12,000/10,000) above 6Vrms = 7.2Vrms.
At 20Hz, the situation is even worse. Saturation will start at 7.2Vrms x (20/60) = 2.4Vrms.
For hybrid use, these transformers work fine because at 300Hz saturation won’t start occurring until 36Vrms which is more than the output of a 100 W – 150W amplifier.
Last edited:
Bolserst: Depressing news, but what I love about this forum - Thank you for sharing your knowledge and saving me lots off frustrating hours in the work shop!
So if I get things right, stripping my 400VA toroid from its original secondary winding - add more turns to the primary (my "ES secondary") lets me lower the usable frequency before saturation accurs?
Roger
So if I get things right, stripping my 400VA toroid from its original secondary winding - add more turns to the primary (my "ES secondary") lets me lower the usable frequency before saturation accurs?
Roger
Bolserst: Depressing news, but what I love about this forum - Thank you for sharing your knowledge and saving me lots off frustrating hours in the work shop!
So if I get things right, stripping my 400VA toroid from its original secondary winding - add more turns to the primary (my "ES secondary") lets me lower the usable frequency before saturation accurs?
It gets confusing talking about primary and secondary of transformers when they mean one thing when the transformer is used as a power transformer and another when used as and ESL step-up transformer.
Winding naming convention:
For power transformer use, lets call the windings by their voltage ratings.
Your 400VA transformer has a 240V winding and a 5V winding.
For ESL use lets call the windings by:
HV=High voltage winding connected to stators
LV=Low voltage winding connected to the driving amplifier.
First, understand that saturation of the core is determined by the winding that is being driven. For power transformer use it is determined by the voltage applied to the 240V winding. For ESL use, it is determined by the voltage applied to the LV winding. So, no matter how many turns you add to the HV winding, the core will saturate at the same input voltage if the number of turns in the LV winding remains the same.
Your question restated:
What I think you are saying is: you want to remove the 5V winding and add more turns to the 240V winding. This would be your HV winding. You would then need to wind on a LV winding to connect your amplifier to. If you increased the number of turns in the LV winding by the same ratio as you increased the number of turns in the HV winding you would still have the same step-up ratio (50:1) and you will have increased the allowable input voltage before the core starts to saturate.
An example:
A typical 400VA toroidal power transformer is wound at about 0.4Volt/turn.
Your transformer may be different, but it will be pretty close.
So, your 240V winding has about 240/0.4 = 600 turns.
And, the 5V winding has about 12 turns.
Lets say you wanted to be able to drive the transformers at 30Hz with a 100Watt amp.
Input to the LV winding would be about 30Vrms so the required turns for the LV winding is roughly 12turns * [ 30V/(5V*1.2) * (60Hz/30Hz)] = 120 turns.
So you need to increase the number of turns in the LV winding by a ratio of 10 from the original 12 in the 5V winding to 120.
To keep the 50:1 step-up ratio you would need to increase the number of turns of the 240V winding by the same ratio, from 600 to 6000.
Adding 5400 turns of wire to a toroidal is not easy to do.
If you chose a full power bandwidth limit of 60Hz instead of 30Hz your LV winding would need to have 60 turns and your HV winding 3000.
bolserst:
I may have totaly confused my measurements, but as I recal I got about 1V/turn on my 400VA trafo!?
With 12 turns on my LV side I get a transformation ratio off abt. 1:20.
This should make things a bit easier as to ad windings to the HV side, but I definitly get what you are saying!
I'm beginning to think that my setup perhaps is good enough as it is!
The bass panel plays with good output from ~30Hz and I get distorsion only when I crank my small 60W amp up to 75%.
Roger
I may have totaly confused my measurements, but as I recal I got about 1V/turn on my 400VA trafo!?
With 12 turns on my LV side I get a transformation ratio off abt. 1:20.
This should make things a bit easier as to ad windings to the HV side, but I definitly get what you are saying!
I'm beginning to think that my setup perhaps is good enough as it is!
The bass panel plays with good output from ~30Hz and I get distorsion only when I crank my small 60W amp up to 75%.
Roger
Are the Dayton-Wright speakers the ones best to win the full-range trophy? Good output below 60 Hz. But even so, there's no magic method for keeping the impedance in a tolerable range down there.
Each transformer (conventional winding, I believe) weighs 41 lbs. They were created by a transformer guru at Hammond Transformers.
I wonder if some of these "interface units" are floating around? They don't get attacked by cats, after all.
I suppose all of us think this is pure Don Quixote hubris, making bass with an ESL. But I gotta say, "WOW, fascinating idea... wonder what it sounds like". Not sure there are serious objections to using cones for sub-woofer and everybody does it. But I sure am curious about the sound. "Fast bass" eh? How many sq meters of film?
Each transformer (conventional winding, I believe) weighs 41 lbs. They were created by a transformer guru at Hammond Transformers.
I wonder if some of these "interface units" are floating around? They don't get attacked by cats, after all.
I suppose all of us think this is pure Don Quixote hubris, making bass with an ESL. But I gotta say, "WOW, fascinating idea... wonder what it sounds like". Not sure there are serious objections to using cones for sub-woofer and everybody does it. But I sure am curious about the sound. "Fast bass" eh? How many sq meters of film?
Last edited:
Hi
Power transformers have a greater “leakage inductance” than say tube output transformers BUT other than that work the same way.
The power transformers I have measured, often were just fine for audio up to 500-1KHz.
All cored transformers have a maximum flux which is permissible, working through the math, one finds this puts a limit on how much voltage per turn can be tolerated AT some frequency.
Thus, if one used a transformer which was designed to have 110 volts across a winding at 60Hz, one knows then that at 30Hz, the maximum Voltage is half that and at 15Hz half of that again.
So, if you know your max input voltage and frequency, you can scale up to find what the input side spec’s should look like.
On the output, you need a pretty large step up ratio so you might look at more of a high voltage power transformer.
Best,
Tom Danley
Power transformers have a greater “leakage inductance” than say tube output transformers BUT other than that work the same way.
The power transformers I have measured, often were just fine for audio up to 500-1KHz.
All cored transformers have a maximum flux which is permissible, working through the math, one finds this puts a limit on how much voltage per turn can be tolerated AT some frequency.
Thus, if one used a transformer which was designed to have 110 volts across a winding at 60Hz, one knows then that at 30Hz, the maximum Voltage is half that and at 15Hz half of that again.
So, if you know your max input voltage and frequency, you can scale up to find what the input side spec’s should look like.
On the output, you need a pretty large step up ratio so you might look at more of a high voltage power transformer.
Best,
Tom Danley
Again this is very true.
I got an incredibly beautiful fullrange sound out of my mini panel at a low level.
But at a more progessive level bass quality quickly diminished.
I last year I was going to get some huge cores but other things came up and now I'll have to wait, hopfully not very long,because I too would like to explore a bass panel setup.
I having been batting around Idea the of a fold panel design with a surface area of at least 32 square feet. jer
I got an incredibly beautiful fullrange sound out of my mini panel at a low level.
But at a more progessive level bass quality quickly diminished.
I last year I was going to get some huge cores but other things came up and now I'll have to wait, hopfully not very long,because I too would like to explore a bass panel setup.
I having been batting around Idea the of a fold panel design with a surface area of at least 32 square feet. jer
You may find it useful to know that most modern toroidal power transformers in the 60VA to 200VA have surprisingly low leakage inductance. Most will drive 500pF - 1000pF ESL panels with an upper bandwidth limit of 25kHz - 40kHz. The only limitation is that of core saturation which restricts their use to hybrids with crossovers at 300Hz and above.
bolserst:
I may have totaly confused my measurements, but as I recal I got about 1V/turn on my 400VA trafo!?
With 12 turns on my LV side I get a transformation ratio off abt. 1:20.
This should make things a bit easier as to ad windings to the HV side, but I definitly get what you are saying!
I'm beginning to think that my setup perhaps is good enough as it is!
The bass panel plays with good output from ~30Hz and I get distorsion only when I crank my small 60W amp up to 75%.
Roger
Sounds like your transformer was designed with a larger core size than the 400VA units I have tested. As you stated, this allows the use of fewer turns on both the HV and LV side and makes things easier for you if you decided to try and add turns to the windings.
On the topic of core saturation and distortion.
Our ears are pretty forgiving of distortion at LF. Especially low order harmonics(2nd & 3rd) which dominate at the onset of core saturation. So you can be pushing your bass transformers into saturation onset and not notice it. If you are hearing audible distortion, you are probably driving the transformer into, or close to, hard saturation.
Transformer core saturation is not an all or nothing event. At the onset of saturation the current in the LV winding stops following the applied voltage. In particular, if you monitor the current, you will notice current spikes on top of the audio signal. As you push further and further into saturation, the magnitude of the spikes grow. Hard saturation is reached when the current is limited only by the DC resistance of the winding and any series resistor added between the LV winding and the amplifier. This is a good reason to always use at least a 1 ohm resistor in series with the LV winding of a bass transformer to protect your amplifier.
The only truly full range ESL I have been able to measure was a Sound Lab A-1. It had solid output below 30Hz. Their interface unit uses a nearly identical setup to Acoustat where they have a large bass transformer with high turns count and high leakage inductance mixed with a toroidal treble transformer with low turns count and low leakage inductance. The toridal transformer is only driven above about 250Hz.
Since this is a thread about ESL Bass transformers, here are a few particulars on the Acoustat and Sound Lab bass transformers.
Acoustat 1+1 bass transformer: step-up ratio(250:1)
EI transformer with core = 3.0 in^2 = 19.4 cm^2
Primary winding = 40 turns
Secondary winding = 10,000 turns
Soundlab A-1 bass transformer: step-up ratio(200:1)
EI transformer with core = 4.5 in^2 = 29 cm^2
Primary winding = 120 turns
Secondary winding = 24,000 turns
Now you can see why I call the Soundlab truly full range.
Onset of saturation is not reached at 30Hz till 50Vrms!
Bolserst wrote “You may find it useful to know that most modern toroidal power transformers in the 60VA to 200VA have surprisingly low leakage inductance. Most will drive 500pF - 1000pF ESL panels with an upper bandwidth limit of 25kHz - 40kHz. The only limitation is that of core saturation which restricts their use to hybrids with crossovers at 300Hz and above.”
That is interesting, all of the ones I had measured were larger and EI cores.
“Now you can see why I call the Soundlab truly full range.
Onset of saturation is not reached at 30Hz till 50Vrms!”
That is a studly transformer.
In American power transformer terms, it would be happy with 100V @60Hz on the primary and would have 20KV on the output.
What you’re looking for here is a high voltage power transformer, not a low voltage unit.
If you want to put bass through it, it has to have a higher voltage primary / input.
Remember what ever the design voltage is at 50 or 60Hz, it will reach the same core flux at half the voltage AT half the frequency.
Search out a Neon sign transformer, oil burner igniter or high voltage power transformer with a 110V primary. These will have more leakage L BUT who cares if it’s a low frequency speaker.
Best,
Tom Danley
That is interesting, all of the ones I had measured were larger and EI cores.
“Now you can see why I call the Soundlab truly full range.
Onset of saturation is not reached at 30Hz till 50Vrms!”
That is a studly transformer.
In American power transformer terms, it would be happy with 100V @60Hz on the primary and would have 20KV on the output.
What you’re looking for here is a high voltage power transformer, not a low voltage unit.
If you want to put bass through it, it has to have a higher voltage primary / input.
Remember what ever the design voltage is at 50 or 60Hz, it will reach the same core flux at half the voltage AT half the frequency.
Search out a Neon sign transformer, oil burner igniter or high voltage power transformer with a 110V primary. These will have more leakage L BUT who cares if it’s a low frequency speaker.
Best,
Tom Danley
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.