Opinions
What are your opinions about those RM14 with air gap, to be used for output coils?
Winding 20 spires I have obtained 80microhenries (80uH)
Also I got litzwire for winding 200 litz of 0,08mm2, Cu-Em and textil material for isolation.
I'm intending to use 2x wires per bobbin.
Some comments please?
I know someone of you, have used those TDK RM 14 cores.
Picture will tell all about litzwire and TDK cores...
What are your opinions about those RM14 with air gap, to be used for output coils?
Winding 20 spires I have obtained 80microhenries (80uH)
Also I got litzwire for winding 200 litz of 0,08mm2, Cu-Em and textil material for isolation.
I'm intending to use 2x wires per bobbin.
Some comments please?
I know someone of you, have used those TDK RM 14 cores.
Picture will tell all about litzwire and TDK cores...
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Coils
Coils realised on TDK cores,80uH, supplied at 94Vcc, I obtained
61,2 VRMS on 4 ohm, when power supply dropping to 87-88Vcc at full load till clipping prag.
The output sinks are a little bit warmer around 50 degrees, after 20 minutes of bench,the coils are COLD(room temperature), and all is OOOOOKKKKK!😉
Pictures and a small film:
Power supply and residual out
Output coils
Same heatsinks
Playing Movie:
http://tinypic.com/player.php?v=2q0ttaw&s=3
Coils realised on TDK cores,80uH, supplied at 94Vcc, I obtained
61,2 VRMS on 4 ohm, when power supply dropping to 87-88Vcc at full load till clipping prag.
The output sinks are a little bit warmer around 50 degrees, after 20 minutes of bench,the coils are COLD(room temperature), and all is OOOOOKKKKK!😉
Pictures and a small film:
Power supply and residual out
An externally hosted image should be here but it was not working when we last tested it.
Output coils
An externally hosted image should be here but it was not working when we last tested it.
Same heatsinks
An externally hosted image should be here but it was not working when we last tested it.
Playing Movie:
http://tinypic.com/player.php?v=2q0ttaw&s=3
O yea, looks like amp has a lot of power, what do you use for power supply?
That big transformer in vid?
That big transformer in vid?
luka said:
Yes, but its so big because I don't have any transformer available, I sold all...
😀
Tomorrow I will play with IRFP260N at more than 100Vcc suply, to see what's hapenning.
The amplifier is ROCK stable, and in idle is eating 16 miliamperes from Vcc, and on his output bridge terminals the tension is lower than 40 milivolts of 125 Khz freq.
Also residual DC out is under 10 milivolts.
I left him 2 hours with couple watts on output and the heatsinks are around 26 degrees, and output coils the same, I have 23 degrees in my home(celsius degrees).
With this values it saturates at 12...15 A. I've already said that 80 uH is too (toooooooo) much.
Otherwise RM14 would be a good core (for a little less power then your planned one), H7C1 is a low-loss material, and 200*0,08 litze is also good, but the air-gap is too small.
Pafi said:
Ok, I will try to lower the inductance (that was in my mind too), and maybe having a bigger air gap.
Also at maximum or minimnum output, the coils stays COLD!
Re: bien
Thank you!
I will try to push-em up with adequate mosfets, coils,heatsinks(I'm a little bastard with the one that I'm used
),and a bigger tension +Vcc.
kanon said:nice job
Thank you!
I will try to push-em up with adequate mosfets, coils,heatsinks(I'm a little bastard with the one that I'm used
),and a bigger tension +Vcc.DiLeco,
Have you posted a schematic of what you are building? If so, where is it? If not, please do so.
Rick
Have you posted a schematic of what you are building? If so, where is it? If not, please do so.
Rick
DjLeco!
If you really want to see how it works, then make a little DC offset on output, load it by a low value resistor (eg. 1 V, 0.05 ohm), and measure everything (current consumption, output residual, Vds waveform, temperatures, etc...)!
About your litz-wire:
Where did you get it from? Is the thicker one still available?
0.04 and 0.08 is the diameter, not cross-section! (only mm, not mm^2)
If you really want to see how it works, then make a little DC offset on output, load it by a low value resistor (eg. 1 V, 0.05 ohm), and measure everything (current consumption, output residual, Vds waveform, temperatures, etc...)!
About your litz-wire:
Where did you get it from? Is the thicker one still available?
0.04 and 0.08 is the diameter, not cross-section! (only mm, not mm^2)
yes
I found somewhere to a buddy of mine, but, is a firm in Germany wich produces alot of variety , http://www.elektrisola.com/en/litz-wire/manufacturing-program.html
Please be more explicit with output coils, I'm only eyes and ears!
Also wich equivalent have my 200x0,08 litzwire in mm2?
I want to try to use 2 wires in parralel to have equivalent of 400x0,08mm,to be sure that is strong enough for section and amperage.
What is your advice in this case?
.Pafi said:
I found somewhere to a buddy of mine, but, is a firm in Germany wich produces alot of variety , http://www.elektrisola.com/en/litz-wire/manufacturing-program.html
Please be more explicit with output coils, I'm only eyes and ears!
Also wich equivalent have my 200x0,08 litzwire in mm2?
I want to try to use 2 wires in parralel to have equivalent of 400x0,08mm,to be sure that is strong enough for section and amperage.
What is your advice in this case?
luka said:
Good Luck, also If is easy for you to buy,I'm interesed to buy me too from you, an mosor (aprox 2-3 Kg) of 300-600-900 lits of 0,08, I tryed to order but they told me, minimum order is aroung 30 kilos(too expensive for me...)
Keep us in touch, I think 600 litzX0.08 is perfectly for anything, or 3000 litz of 0,04 is enough for 1Mhz amplifier.
Any other opinions?
Luka!
Please make sure you really can buy litz wire!
DjLeco!
I think 600*0,08 would be too thick, very difficult to handle in many cases. But I'm afraid we actually don't have the option to choose whatever we want.
Supposing it is really available in small quantity, the price is still a question. If it costs that much as I remember, then I stay at making it for myself.
I referred to the whole amp, but yes, the output coil also can be checked this way. (However if you particularly want to check output coil against saturation, then it's better to do with relatively short pulses, by seeing residual on scope.)
Is this a serious question? Cross-section (=area) of circle is A=d^2*pi/4 = 0.08*0.08*3,14/4=0.005 mm^2 (=> total area of 1 mm^2).
About choosing the right cross-section: if you use litze, and core doesn't saturate, the power-loss comes from core loss (wich is very low), and ohmic resistance of wire. Eg. if wire-length is 1 meter, and cross-sectional area is 2 mm^2, then resistance is approximately 0.0085 ohm, so 20 A makes Ploss=0.0085*20*20=3.4 W. This power is near the limit wich can be dissipated safely by the surface of an RM14 without forced cooling, but it's OK, because it have to carry only the effective (average) current of the speakers, wich is usually much lower then peak current. (Typically long-term effective value is less then 1/4 th of the peak current.)
But core have to be able to withstand peak current! After the core saturates, eddy-current goes up, producing much more heat. Nevertheless, saturation makes distortion also. Saturation occures when amperes*turns exceeds d*Bmax/mu0, where d=height of air-gap (in meters!), Bmax= saturation inductance (~0.27...0.4 T), mu0 is permeability of air (~1,26*10^-6 Vs/Am).
Gap height shouldn't be more then ~2 mm, because over this (approximate) limit flux leakage becomes too high, eddy-current starts to be significant.
Please make sure you really can buy litz wire!
DjLeco!
I think 600*0,08 would be too thick, very difficult to handle in many cases. But I'm afraid we actually don't have the option to choose whatever we want.
Supposing it is really available in small quantity, the price is still a question. If it costs that much as I remember, then I stay at making it for myself.
I referred to the whole amp, but yes, the output coil also can be checked this way. (However if you particularly want to check output coil against saturation, then it's better to do with relatively short pulses, by seeing residual on scope.)
Is this a serious question? Cross-section (=area) of circle is A=d^2*pi/4 = 0.08*0.08*3,14/4=0.005 mm^2 (=> total area of 1 mm^2).
About choosing the right cross-section: if you use litze, and core doesn't saturate, the power-loss comes from core loss (wich is very low), and ohmic resistance of wire. Eg. if wire-length is 1 meter, and cross-sectional area is 2 mm^2, then resistance is approximately 0.0085 ohm, so 20 A makes Ploss=0.0085*20*20=3.4 W. This power is near the limit wich can be dissipated safely by the surface of an RM14 without forced cooling, but it's OK, because it have to carry only the effective (average) current of the speakers, wich is usually much lower then peak current. (Typically long-term effective value is less then 1/4 th of the peak current.)
But core have to be able to withstand peak current! After the core saturates, eddy-current goes up, producing much more heat. Nevertheless, saturation makes distortion also. Saturation occures when amperes*turns exceeds d*Bmax/mu0, where d=height of air-gap (in meters!), Bmax= saturation inductance (~0.27...0.4 T), mu0 is permeability of air (~1,26*10^-6 Vs/Am).
Gap height shouldn't be more then ~2 mm, because over this (approximate) limit flux leakage becomes too high, eddy-current starts to be significant.
...
The airgap of my cores is exactlly 2mm!
If I have good understand, the equivalent of my 200x0.08 lits is around 1mm in normal wire?
The lenght of wire winding is exactly 1 meter.
Also, wich is maximum current for my 200X0.08 litz?
Is OK to wind with 2 of it in parralel for higher current and minim resistance, or is enough just one?
I'm not so familiar with mathematics(never liked,shame me!)
The airgap of my cores is exactlly 2mm!
If I have good understand, the equivalent of my 200x0.08 lits is around 1mm in normal wire?
The lenght of wire winding is exactly 1 meter.
Also, wich is maximum current for my 200X0.08 litz?
Is OK to wind with 2 of it in parralel for higher current and minim resistance, or is enough just one?
I'm not so familiar with mathematics(never liked,shame me!)
How is it possible? On the picture it seems to be much smaller, and by calculating back from inductance and number of turns it comes to be 1 mm also!
1 mm^2. It equals to 1.13 mm in diameter.
For long time average, without forced cooling, in an RM14 core, it's about 10 A (since I said 2 mm^2 can carry 20 A), however the (straight) wire in free air can carry up to 20 A without damage.
This was in my example. But if you ask about "higher" current, it would be advisable to tell me how much is the "lower" current (you could also specify the "higher" one)! These words doesn't mean anything without specifying the reference.
Pafi said:
Telling I want to use those output coils for 2 ohms and 4 ohms speakers,presuming I gave around 100V RMS at output, suplying the power module with 140Vcc 20Amps.
See the pictures with new air gaps dimensions, and the tool of polishing 'em under water.
You say 2mm air gap is enough, or need more?
Is there a problem that the polished surfaces are NOT perfectly planes like the original gap?
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- Not open for further replies.