Hi,
I've made some tweeter ribbons with an impedance of 1Ohm.
I like to avoid transformer, so want to check if there's some easy to build amplifiers out there?
I assume the voltage doesn't have to be that High.
Actually I was planning to use a opamp with a single driver stage behind. But want's to check for other ideas.
I assume a current amplifer would also make a good job here. The crossover is done in the digital domain, so no need to consider this.
I've made some tweeter ribbons with an impedance of 1Ohm.
I like to avoid transformer, so want to check if there's some easy to build amplifiers out there?
I assume the voltage doesn't have to be that High.
Actually I was planning to use a opamp with a single driver stage behind. But want's to check for other ideas.
I assume a current amplifer would also make a good job here. The crossover is done in the digital domain, so no need to consider this.
Here is something that I came up with the other day,
http://www.diyaudio.com/forums/solid-state/167394-quasi-amplifier-beginners-10.html#post2812503
Although I haven't built it yet,it does simulate pretty good.
jer
http://www.diyaudio.com/forums/solid-state/167394-quasi-amplifier-beginners-10.html#post2812503
Although I haven't built it yet,it does simulate pretty good.
jer
Yes very much so at 1 ohm as the current out will match the voltage out.
I haven't tried it yet in the sim,But logic tells me to double up on the output devices for a 1 ohm load.
Although a 66Vp-p swing is a lot of power at 544.5 watts into 1 ohm.
I was also able to set the bias pretty high for class A operation as well,as it is just a sim for now.
I do plan on building this one some time.
jer
I haven't tried it yet in the sim,But logic tells me to double up on the output devices for a 1 ohm load.
Although a 66Vp-p swing is a lot of power at 544.5 watts into 1 ohm.
I was also able to set the bias pretty high for class A operation as well,as it is just a sim for now.
I do plan on building this one some time.
jer
Hello,
Read through this for a start ...
http://www.diyaudio.com/forums/pass-labs/8353-apogee-scintilla-amp.html
opamp ...
Read through this for a start ...
http://www.diyaudio.com/forums/pass-labs/8353-apogee-scintilla-amp.html
opamp ...
Hi,
Thanks for all the posts.
I would really prefer if it's not a classA, since I want a build in amplifer. I guess the powersupply could be placed seperatly.
Another important parameter is that there should be no, or very little voltage offset at the output.
Seems like opamp with buffer, is not seen as good solution, any specific reason for it?
but anyway, it's only needed to have a gain of 1x if a poweramplifier is used. Any good idea of stage to add to the poweramplifier in order to reduce the load, seen from amplifier?
Thanks for all the posts.
I would really prefer if it's not a classA, since I want a build in amplifer. I guess the powersupply could be placed seperatly.
Another important parameter is that there should be no, or very little voltage offset at the output.
Seems like opamp with buffer, is not seen as good solution, any specific reason for it?
but anyway, it's only needed to have a gain of 1x if a poweramplifier is used. Any good idea of stage to add to the poweramplifier in order to reduce the load, seen from amplifier?
Sounds OK to me, as long as you're not of the "all opamps sound lousy" mindset.
At least it's a (relatively) easy way to get something up and running. You can always swap the "opamp" part of it for something discrete later on if you feel the need, and keep the buffer part of the original.
After all, the big money normally goes to the power supply, heatsinking etc. The small-signal stuff is relatively inexpensive.
One thing though: I would use a separately filtered/regulated supply for the opamp as the output buffer is likely to put a lot of ripple on it's supply rails.
There was some discussion about this on Audiocircle. At least one of the designs
used the IXTH20n50 the same device used in the Zenlite but used it in a self biasing
circulotron output stage.
Two simple Current Amplifiers for low impedance loads
used the IXTH20n50 the same device used in the Zenlite but used it in a self biasing
circulotron output stage.
Two simple Current Amplifiers for low impedance loads
The only way that I can think of to change the impedence load seen by the amplifier is to use a transformer.
This has been discussed before in great detail in another thread.
A 2 to 1 winding ratio will allow the amplifier to see your 1ohm load as a 4ohm load.
And because it is for higher frequency's it would not require a large core or very many turns.
It can be done ,but designing an amp for a 1 ohm load (or lower) always seems to have its issues.
A design that I think that has been done but not well documented and one that I would like to try as well is to parallel about 8 chipamps either the Lm3886 or some LM4870's.
With each section with its own DC Servo circuit so that it will stay stable and not be a PIA to adjust the offset of each individual power chip.
The main question is how much current can the ribbon handle before it disintegrates, then you can start to design a power amp using only the maximum amount of voltage you need for the power supply so that you don't fuse the ribbon.
Just I few more idea's.
jer
yes
and besides, any theoretical loss caused by a transformer would probably be much less 'degrading' than the compromised result from any 'cheaply' designed electronics capeable of 1ohm load
another advantage of the transformer could be that it might deal with the usual top end hash
Put it this way we use transformers on our DIY ESL's and they are the most incredible sounding things I have ever heard.
Just start with a good quality core.
The few THD measurements I did do on my cores were way below .1% and much more towards .01% range through out it's operating range at around 300Hz to 400Hz and above.
I don't remember the exact numbers,But I think I was just seeing the THD of the amplifier itself through the transformer.
Low frequency's and core saturation are the biggest factor to be worried about.
Also try to get the secondary winding DC resistance as low as you can, as well as complete core coverage to to keep the leakage inductance at a minimum if you choose a Torrid type core.
There is a thread that I started the discusses these things but I drop the project for lack of interest as I am mostly into ESL's.
But I do like to mess with ribbons as well some times.
There is a thread buried some where that discusses a complete design and some core examples for ribbons I will look for it later and post the link for you,If you like.
jer
Just start with a good quality core.
The few THD measurements I did do on my cores were way below .1% and much more towards .01% range through out it's operating range at around 300Hz to 400Hz and above.
I don't remember the exact numbers,But I think I was just seeing the THD of the amplifier itself through the transformer.
Low frequency's and core saturation are the biggest factor to be worried about.
Also try to get the secondary winding DC resistance as low as you can, as well as complete core coverage to to keep the leakage inductance at a minimum if you choose a Torrid type core.
There is a thread that I started the discusses these things but I drop the project for lack of interest as I am mostly into ESL's.
But I do like to mess with ribbons as well some times.
There is a thread buried some where that discusses a complete design and some core examples for ribbons I will look for it later and post the link for you,If you like.
jer
here is one of them,
http://www.diyaudio.com/forums/planars-exotics/50162-another-diy-ribbon-thread.html#post558592
another great spot in the same thread,
http://www.diyaudio.com/forums/planars-exotics/50162-another-diy-ribbon-thread-11.html#post2039651
here is another one,
http://www.diyaudio.com/forums/plan...-tl-loaded-extremis-hybrid-4.html#post1527698
http://www.diyaudio.com/forums/plan...ns-transformer-cores-ribbons.html#post2676930
http://www.diyaudio.com/forums/cons...nding-1st-toroid-transformer.html#post2510958
This is the one were I started a go at it then I lost interest due to the last post as my intent was not to take over the thread as I had a genuine interest and I felt that the last post was a bit rude and could have been sent in a PM ,but that's the way it goes,oh well,I never found out what the statement was all about and I never asked.
http://www.diyaudio.com/forums/plan...d-transformer-ribbon-tweeter.html#post2484880
I re found all of these links by doing a search with "ribbon transformer" and there is a whole lot more cool stuff to read.
http://www.diyaudio.com/forums/tube...-umccc-125-what-would-you-do.html#post2505396
http://www.diyaudio.com/forums/parts/173994-hand-wound-copper-foil-transformer.html#post2309461
http://www.diyaudio.com/forums/tube...bon-speaker-directly-why-not.html#post2264246
http://www.diyaudio.com/forums/planars-exotics/172569-alu-vs-laminated-ribbon.html#post2296107
http://www.diyaudio.com/forums/planars-exotics/170076-how-calculate-inductance-foil.html#post2240749
http://www.diyaudio.com/forums/plan...o-my-ribbion-speaker-project.html#post1879615
this is an interesting thread and multiple paralleled chipamps get discussed ther is another thread where Sebastion post a schematic and PC Board design for a very large paralelled amp and I believe that he had it running as well.
I will try to find the link to that thread as it was just last summer or spring.
http://www.diyaudio.com/forums/chip-amps/164634-chip-amp-rival-hi-end-design-advice.html#post2144636
I hope these help to get you started.
jer
http://www.diyaudio.com/forums/planars-exotics/50162-another-diy-ribbon-thread.html#post558592
another great spot in the same thread,
http://www.diyaudio.com/forums/planars-exotics/50162-another-diy-ribbon-thread-11.html#post2039651
here is another one,
http://www.diyaudio.com/forums/plan...-tl-loaded-extremis-hybrid-4.html#post1527698
http://www.diyaudio.com/forums/plan...ns-transformer-cores-ribbons.html#post2676930
http://www.diyaudio.com/forums/cons...nding-1st-toroid-transformer.html#post2510958
This is the one were I started a go at it then I lost interest due to the last post as my intent was not to take over the thread as I had a genuine interest and I felt that the last post was a bit rude and could have been sent in a PM ,but that's the way it goes,oh well,I never found out what the statement was all about and I never asked.
http://www.diyaudio.com/forums/plan...d-transformer-ribbon-tweeter.html#post2484880
I re found all of these links by doing a search with "ribbon transformer" and there is a whole lot more cool stuff to read.
http://www.diyaudio.com/forums/tube...-umccc-125-what-would-you-do.html#post2505396
http://www.diyaudio.com/forums/parts/173994-hand-wound-copper-foil-transformer.html#post2309461
http://www.diyaudio.com/forums/tube...bon-speaker-directly-why-not.html#post2264246
http://www.diyaudio.com/forums/planars-exotics/172569-alu-vs-laminated-ribbon.html#post2296107
http://www.diyaudio.com/forums/planars-exotics/170076-how-calculate-inductance-foil.html#post2240749
http://www.diyaudio.com/forums/plan...o-my-ribbion-speaker-project.html#post1879615
this is an interesting thread and multiple paralleled chipamps get discussed ther is another thread where Sebastion post a schematic and PC Board design for a very large paralelled amp and I believe that he had it running as well.
I will try to find the link to that thread as it was just last summer or spring.
http://www.diyaudio.com/forums/chip-amps/164634-chip-amp-rival-hi-end-design-advice.html#post2144636
I hope these help to get you started.
jer
Last edited:
the biggest design headache for an ESL is the transformer. Getting the transformer to work well at all bass frequencies and work well at all mid frequencies and to work well at all treble frequencies is a very difficult and time consuming and expensive design and develop exercise.
The Quad els57 is a 3 element ESL.
Would it be feasible to design a bass transformer, a mid transformer and a treble transformer and connect these dedicated transformers to their respective ESL elements?
Each transformer could be driven by it's own dedicated and specialised narrow band amplifier.
I suspect that much of what we hear from an ESL is determined by the transformer's abilities. If we stress the transformer less by asking each of the separated transformers to operate in narrow band, can we expect to hear "more" from the speaker and less from the transformer/s?
Could the same narrowband philosophy be applied to a ribbon? Would the transformer be less "heard"?
The Quad els57 is a 3 element ESL.
Would it be feasible to design a bass transformer, a mid transformer and a treble transformer and connect these dedicated transformers to their respective ESL elements?
Each transformer could be driven by it's own dedicated and specialised narrow band amplifier.
I suspect that much of what we hear from an ESL is determined by the transformer's abilities. If we stress the transformer less by asking each of the separated transformers to operate in narrow band, can we expect to hear "more" from the speaker and less from the transformer/s?
Could the same narrowband philosophy be applied to a ribbon? Would the transformer be less "heard"?
Actually it is the higher frequency's that are a PIA because the the added parasitic capacitance the creates an unneeded extra load on the amplifier.
Along with the leakage inductance that determines what the resonate frequency of transformer and this needs to as far as you can get and it above the audio band and higher if you plan to have high frequency square wave performance.
Core losses due have some impact on high frequency response as well.
This is why your typical large I-E power transformers don't fair well for audio devices.
Not so much because of the cheap iron that is used but because of the way they were wound.
From what I understand is that a lot guitar amps used the cheap iron for OPT's and is what helps to get their characteristic sound.
But this was done mainly because of cost back in the day.
Most today's cores use a higher quality iron rather it be E-I or Toroid cores, as the name of today's game is higher efficiency.
This is not so much of a factor for a step-down transformer and leakage inductance becomes the important factor,but some what is easier to deal with.
Low frequency performance is solely based on how much iron you have and how many turns is in the primary as per peak voltage into the primary.
As you lower the lower the frequency or increase the voltage at above a certain point at the lowest frequency then the core starts to go into saturation.
This is the reason that 50Hz transformers are bigger and why you can use them at 60Hz but not vice versa (The other way around).
Also it is the reason you cannot use a 120v winding on a 220v source but you can use a 220v winding on a 120v source.
To do either of these things you must increase the number of primary turns for the same size core or increase the core area for the same number of primary turns.
Some more good stuff,
http://www.diyaudio.com/forums/tubes-valves/193508-transformer-design-equations.html#post2653389
http://www.diyaudio.com/forums/power-supplies/77919-measuring-inductors.html#post895799
http://www.edn.com/contents/images/6294161.pdf
AndrewT,I had ran across a link of a thread of measuring core saturation it may be buried in the in the links that I had already posted but when I find it again I will repost it for you as it is fairly simple to do.
The above reference is pretty good to although I just use a resistor instead of a capacitor as I described (maybe poorly) in the step-up transformer design thread.
But the trick is to have a signal generator that can supply enough current to be able to make the core saturate.
It wasn't until I had hooked an amplifier up to the core that I could get those measurements.
Jer
Along with the leakage inductance that determines what the resonate frequency of transformer and this needs to as far as you can get and it above the audio band and higher if you plan to have high frequency square wave performance.
Core losses due have some impact on high frequency response as well.
This is why your typical large I-E power transformers don't fair well for audio devices.
Not so much because of the cheap iron that is used but because of the way they were wound.
From what I understand is that a lot guitar amps used the cheap iron for OPT's and is what helps to get their characteristic sound.
But this was done mainly because of cost back in the day.
Most today's cores use a higher quality iron rather it be E-I or Toroid cores, as the name of today's game is higher efficiency.
This is not so much of a factor for a step-down transformer and leakage inductance becomes the important factor,but some what is easier to deal with.
Low frequency performance is solely based on how much iron you have and how many turns is in the primary as per peak voltage into the primary.
As you lower the lower the frequency or increase the voltage at above a certain point at the lowest frequency then the core starts to go into saturation.
This is the reason that 50Hz transformers are bigger and why you can use them at 60Hz but not vice versa (The other way around).
Also it is the reason you cannot use a 120v winding on a 220v source but you can use a 220v winding on a 120v source.
To do either of these things you must increase the number of primary turns for the same size core or increase the core area for the same number of primary turns.
Some more good stuff,
http://www.diyaudio.com/forums/tubes-valves/193508-transformer-design-equations.html#post2653389
http://www.diyaudio.com/forums/power-supplies/77919-measuring-inductors.html#post895799
http://www.edn.com/contents/images/6294161.pdf
AndrewT,I had ran across a link of a thread of measuring core saturation it may be buried in the in the links that I had already posted but when I find it again I will repost it for you as it is fairly simple to do.
The above reference is pretty good to although I just use a resistor instead of a capacitor as I described (maybe poorly) in the step-up transformer design thread.
But the trick is to have a signal generator that can supply enough current to be able to make the core saturate.
It wasn't until I had hooked an amplifier up to the core that I could get those measurements.
Jer
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.