Hello
I have some speakers rated at 4 ohms with a minimum 3.5 ohms at 190hz, that is why i want to bias my amplifier at 1.6-2amp per fet depending on heatsink temperature.
I will build it in two monoblocks each monoblock having 2 x 0.31C/W heatsinks , one 400VA(2 x 19.5V/10A) toroid and an CRC filter 66kuf - 0.22-0.47R - 66kuf
From what i have understood i have to lower R16-R19 and R20-R23 to same value.Would 0.33ohms be enough or i have to lower the value to 0.27R?
What about R27? What value should this have? If want to remove definitely R27 what value do you recommend for R16-R19 and R20-R23 to have so i can bias the amplifier in the desired range?
I am thinking in lowering a little bit the raills to 22-23V so i can manage easily greater bias without having dissipation problems
And finally some first design pictures
(for the moment i have only the heatsinks ,toroids,2SJ109 and matched irfp240,have to buy the rest)
I have some speakers rated at 4 ohms with a minimum 3.5 ohms at 190hz, that is why i want to bias my amplifier at 1.6-2amp per fet depending on heatsink temperature.
I will build it in two monoblocks each monoblock having 2 x 0.31C/W heatsinks , one 400VA(2 x 19.5V/10A) toroid and an CRC filter 66kuf - 0.22-0.47R - 66kuf
From what i have understood i have to lower R16-R19 and R20-R23 to same value.Would 0.33ohms be enough or i have to lower the value to 0.27R?
What about R27? What value should this have? If want to remove definitely R27 what value do you recommend for R16-R19 and R20-R23 to have so i can bias the amplifier in the desired range?
I am thinking in lowering a little bit the raills to 22-23V so i can manage easily greater bias without having dissipation problems
And finally some first design pictures
(for the moment i have only the heatsinks ,toroids,2SJ109 and matched irfp240,have to buy the rest)Attachments
Last edited:
With higher bias i think i will get lower distorsions at 4 ohms, beside that i read few suggestions recommending to double the bias for 4ohms speakers.
My speakers are only 88 db so I think more power won't hurt them much
I am kind of newbie and my knowledge is limited in this domain that is why I am asking for your help so I could finish and enjoy my amplifiers like the rest of you.
My speakers are only 88 db so I think more power won't hurt them much
I am kind of newbie and my knowledge is limited in this domain that is why I am asking for your help so I could finish and enjoy my amplifiers like the rest of you.
Last edited:
First item is that you may have doubled the heatsink capacity, but the larger part of the thermal resistance is at/within the output device.
Going from 100W dissipation to ~180W (2A) by 4 output devices may result in similar heatsink temperature, but output device load goes from 1/6th power and 50% temperature range, to ~1/3d power and ~75% (~115C die temperature)
Second one, even 2A bias will not enable to reach 100dB at 190Hz.
(Plenty 4 ohm loudspeakers that do >6 ohm below and above 1KHz. Driving full Class A at 190Hz midbass is a luxury, similar to taking a full-size 4WD truck to a parking garage of a shopping mall)
Going from 100W dissipation to ~180W (2A) by 4 output devices may result in similar heatsink temperature, but output device load goes from 1/6th power and 50% temperature range, to ~1/3d power and ~75% (~115C die temperature)
Second one, even 2A bias will not enable to reach 100dB at 190Hz.
(Plenty 4 ohm loudspeakers that do >6 ohm below and above 1KHz. Driving full Class A at 190Hz midbass is a luxury, similar to taking a full-size 4WD truck to a parking garage of a shopping mall)
Last edited:
I have found somewhere here on the forum an Microsoft Office Excel document(i atached this to the post) that helps me in calculating Aleph output power at different speaker impedances and shows mosfet junction and heatsink temperature.So to keep junction temperature below 100C i have selected 1.6A per fet @ 24V(this is why i want to lower a little bit power rails from 24 to 22-23) and i get this results : maximum power - 63W@3.5ohm, junction temp - 96C(at this temperature each fet will suport a maximum dissipation of 65W ), heatsink temp - 53C and not last dissipation per fet - 38.4W
Did a little research now on the spl and I found this site Peak SPL Calculator
When you say i can't reach 100db at 190hz what I understand is that i can't reach 10W at 190hz, is that right?
Not really encouraging news but if the sound is good who will care about power?
Did a little research now on the spl and I found this site Peak SPL Calculator
When you say i can't reach 100db at 190hz what I understand is that i can't reach 10W at 190hz, is that right?
Not really encouraging news
but if the sound is good who will care about power?
Standard Aleph-J does 2.5A peak, twice the bias level makes ~5A (is for 2A per device, not 1.6A).
5 Amps times 3.5 Ohm is 17.5V peak, is < 12.5V continuous.
12.5V makes less than 20W in an 8 ohm load, is about 13dB up from 2.83V or 1W/8.
(from 20W to 200W is +10dB, from 200W to 100W is -3dB. => 20W is 7dB down from 100W. => 100W is +20dBW => 20W is +13dB. Me like head counting
)
If your loudspeakers are 88dB/2.83Vac/M => 88 +13 = 101dB SPL.
Unfortunately, there's a thing called phase shift, voltage out of phase with current, so you'll likely not reach the 100dB level for 190Hz.
Try using the correct number at line 7 of William's calculator, instead of 50%.
Typing is nice, especially with cute fingernails, using head better.
(me chopped half an inch of my left middle finger off with a milling machine yesterday, so typing is out for a while)
5 Amps times 3.5 Ohm is 17.5V peak, is < 12.5V continuous.
12.5V makes less than 20W in an 8 ohm load, is about 13dB up from 2.83V or 1W/8.
(from 20W to 200W is +10dB, from 200W to 100W is -3dB. => 20W is 7dB down from 100W. => 100W is +20dBW => 20W is +13dB. Me like head counting
)If your loudspeakers are 88dB/2.83Vac/M => 88 +13 = 101dB SPL.
Unfortunately, there's a thing called phase shift, voltage out of phase with current, so you'll likely not reach the 100dB level for 190Hz.
Try using the correct number at line 7 of William's calculator, instead of 50%.
Typing is nice, especially with cute fingernails, using head better.
(me chopped half an inch of my left middle finger off with a milling machine yesterday, so typing is out for a while)
Last edited:
I do not know what number to use there because i do not know how to calculate ac gain, it's too much for my knowledge.From what I have understood ac gain is 50% in standard schematic, and that is the value i want to use.So I guess if increase bias and I lower R16-R19 and R20-R23 I have to lower R24 too to maintain ac gain at 50%.But I still don't know what values I should use for R16-R19 and R20-R23 for 1.6-2A bias per fet Can anybody help?
Thank you for your answers and sorry for your finger
Can anybody help?Thank you for your answers and sorry for your finger
As you said, 0R33 for R16 to R23.
At 1.6A, voltage drop across the source resistor will be a little above 500mV.
Vgs will be higher for 1.6A, easiest to adjust is by swapping in trimpots for the DC bias resistors, get a reading, then exchange for a same value resistor.
(btw, sensitivity of your loudspeakers is 86dB/2.83V/m, as you posted earlier. At 3.2A bias total you'll likely not even reach 97dB SPL at 190Hz. And looking at the impedance curve, even much less. Seems a whole lot of heat for little volume. Just mentioning it. http://www.soundstagenetwork.com/measurements/speakers/dynaudio_excite_x12/ )
At 1.6A, voltage drop across the source resistor will be a little above 500mV.
Vgs will be higher for 1.6A, easiest to adjust is by swapping in trimpots for the DC bias resistors, get a reading, then exchange for a same value resistor.
(btw, sensitivity of your loudspeakers is 86dB/2.83V/m, as you posted earlier. At 3.2A bias total you'll likely not even reach 97dB SPL at 190Hz. And looking at the impedance curve, even much less. Seems a whole lot of heat for little volume. Just mentioning it. http://www.soundstagenetwork.com/measurements/speakers/dynaudio_excite_x12/ )
Last edited:
Well, i stopped smoking.
So, if i do this for 9 more days, i can't smoke, even if i wanted to.
Not much finger to find with a wood milling machine running at a zillion rpm.
I give all doctors the finger, and lawyers.
(I'm an asparagus, which is just as well after +40 years of handling heavy machinery)
So, if i do this for 9 more days, i can't smoke, even if i wanted to.
Not much finger to find with a wood milling machine running at a zillion rpm.
I give all doctors the finger, and lawyers.
(I'm an asparagus, which is just as well after +40 years of handling heavy machinery)
Last edited:
Well, i stopped smoking.
So, if i do this for 9 more days, i can't smoke, even if i wanted to.
Not much finger to find with a wood milling machine running at a zillion rpm.
I give all doctors the finger, and lawyers.
(I'm an asparagus, which is just as well after +40 years of handling heavy machinery)
Damn... I hear women like asparagus.
Hello
I want to order 4 thermal circuit breakers,I found two models but i don't know which one to choose from them.
They are these ones :
http://www.adelaida.ro/product_info.php?products_id=10577&osCsid=8926c99eedf4e1d97d4d7faebe271135
and
Termostat bimetalic BT-H70H - ADELAIDA - Componente electronice
The only thing differing at them is the closing type which is horizontal or vertical.I would like to mount them like in this link http://www.rstaudio.de/AlephJ/ajKuehlkoerper.jpg so which one should i choose?
One more question : for 1.6-1.7A bias per fet what fuse should I use on mains? 0.8-1A fast or slow blow would do the job?
Thank You
I want to order 4 thermal circuit breakers,I found two models but i don't know which one to choose from them.
They are these ones :
http://www.adelaida.ro/product_info.php?products_id=10577&osCsid=8926c99eedf4e1d97d4d7faebe271135
and
Termostat bimetalic BT-H70H - ADELAIDA - Componente electronice
The only thing differing at them is the closing type which is horizontal or vertical.I would like to mount them like in this link http://www.rstaudio.de/AlephJ/ajKuehlkoerper.jpg so which one should i choose?
One more question : for 1.6-1.7A bias per fet what fuse should I use on mains? 0.8-1A fast or slow blow would do the job?
Thank You
Last edited:
Horizontal & Vertical merely refers to the orientation of the push-on terminals, whatever you find more practical or prettier.
Fast blow on mains ? Why ? (think peak currents at turn-on, transformer and capacitors)
1A times 230V = 230VA. Is that smart engineering if the transformer behind the fuse is 400VA ?
(when the end comes i know, i was just an asparagus, life goes on without me)
Fast blow on mains ? Why ? (think peak currents at turn-on, transformer and capacitors)
1A times 230V = 230VA. Is that smart engineering if the transformer behind the fuse is 400VA ?
(when the end comes i know, i was just an asparagus, life goes on without me)
Attachments
Last edited:
I think i need slow blow type fuse but i do not know how to calculate the rated current for the fuse
pretty ugly finger you have there
Last edited: