Where were you contemplating putting these resistors? Each module of the L20D is capable of driving 3 ohms, but they have a maximum average output current capability of about 6 amps (at 70C). That would mean approx (6 / 1.414) x (6 / 1.414) x 3 = 54watts maximum into 3 ohms, and you would want to keep the supply voltage at a minimum, say 50-0-50 volts or so, (so as to reduce dissipation in the IRFI4020). Is 54 watts enough for you?
There may be amplifiers more suitable for that 3 ohm load. Have you looked at the T2, T3 and T4 from HifimeDIY? Unfortunately their website seems to be down at this moment. But there are plenty of discussion threads here on diyAudio
.
Attachments
We can not violate the principle of Ohm's law.
L20D in the case of 8 ohms (about +-63V DC) power can output 200W, 5A approximate.
However, in 4 European time, to achieve the same power, the current takes about to 7A, the voltage only need about +-50V.
The BTL can not increase the output current, which can increase the output voltage.
I suggest you reduce the transformer voltage, and improve the current transformer.
With L15D or L15DX2, since they are original in IRAUDAMP7S-150. In the case of 4 ohms 250W power output (+-50V)
Very correct
Lower impedance, the same power needs of a larger current, which has a huge MOSFET ID requirements, and output inductor are also a great test.
But it needs the power supply voltage also reduces the requirements.
In fact the most important thing is we know we need power.
Thanks for replying.
I'm thinking about tying 2 outputs per channel with resistors. I have a +/-70V PS from a Soundcraftsmen PM860 (8.5A). It provides plenty of current. I have run a pair of PM860s bridged at 4 ohms pushing subwoofers at a party without problems.
I don't have a +/-50V PS.
Whatever amplifier module you choose will have significant excess heat dissipation from such a high supply. The L20D should be able to handle it OK, but few of the others will.
Ljm_ljm warned the inductor might saturate, but the 1D17A-220M inductor (clone) used in the L20D is capable of handling 6 Amps average, from a heat perspective, while its peak (saturating) current is way higher, at about 15A.
So the key element is the output transistor, the IRFI4020-117.
If it is a clear day, and the wind is blowing in the right direction, and you keep the heatsink really cool, this part is rated at 9 Amps continuous, 36 Amps peak. The 9 average amps falls to 6 Amps as the transistor heats to 100 degrees C. If you decide to push the L20D to its limits, you will get about 120W continuously out of each module.
I guess it is possible to bridge two modules using high-power resistors, but, as a designer, this is not something I would want to do. There are many complex issues involved, including feedback loop stability.
The first thing I would do is measure the actual impedance of the load (subwoofer?). I most cases it will be higher than 3 ohms around the enclosure resonances, which is where most of the power probably lies. If this is the case then one L20D will be able to safely drive the 120W average level, with occasional peaks easily reaching 250W.
Note: YMMV - Your mileage may vary - I haven't tested the L20D below 4 ohms.
.
Thanks trevmar,
I'm ok with dissipation. I'm actually pulling the output board from the amplifier which is fan cooled and dropping these in. I got to figure how insert the temperature sensing transistors into the L20D heat sinks. I will follow your advise and use one per channel and see what I get.
I'm ok with dissipation. I'm actually pulling the output board from the amplifier which is fan cooled and dropping these in. I got to figure how insert the temperature sensing transistors into the L20D heat sinks. I will follow your advise and use one per channel and see what I get.
Add a small fan active ventilation devices, reducing the MOSFET and the inductor temperature, you can get more power margin range.
This is a very simple way, can be installed on top of the radiator, while reducing the heat sink MOSFET, and the inductor temperature.
But may bring some fan noise.
This is a very simple way, can be installed on top of the radiator, while reducing the heat sink MOSFET, and the inductor temperature.
But may bring some fan noise.
@rickedd46 , else you need to wait till someone design the IRAUDAMP9 these are capable to drive 2ohm with your trafo
IRAUDAMP9 in 2 when Europe can reach 1000 W of power.
You have to go to embrace a great transformer, about 20 KG 2000 W of transformer!!!!!
This is more like weapons not Musical Instruments.
I think most people are actually using less than, really 2's speaker is a few exist.
If you don't mind, analog amplifier actually also is right choice.
For example L150W
http://www.diyaudio.com/forums/solid-state/191738-new-design-l-150w-fet-amplifier.html
I have not tried L20D in 2 European work what circumstance,
Should be able to work only output current limiting, power could be less than 8 the load.
Should not be more than 100 W.
Last edited:
L20D - PSU.. how low can you go?
Although the board has printed +-60V < IN < +-80v on it seems you are both testing (and I presume that also includes listening) with lower voltages with no issue.
I have 8 ohm, fairly high efficiency speakers and dont need anywhere near the maximum power these amps use: so I should be good running them with a lower DC supply?
From some of the comments it seems they may measure better that way?
One idea I am considering is making monoblock amps, each powered by a single +-54V DC ConnexElectronics A350SMPS.
Though it looks like even using one of these to power both might give me a lot of amp for the money?
Incidentally, my two boards only needed about 45 volts to start operating properly, and I haven't taken them above 55V yet. They are running beautifully there, giving my 100W 4/6/8 dummy-load a run for its money at that voltage
Although the board has printed +-60V < IN < +-80v on it seems you are both testing (and I presume that also includes listening) with lower voltages with no issue.
I have 8 ohm, fairly high efficiency speakers and dont need anywhere near the maximum power these amps use: so I should be good running them with a lower DC supply?
From some of the comments it seems they may measure better that way?
One idea I am considering is making monoblock amps, each powered by a single +-54V DC ConnexElectronics A350SMPS.
Though it looks like even using one of these to power both might give me a lot of amp for the money?
Three resistors cause the shutdown at around 40-45V. These are 1W resistors used to feed the zener diode (5V) regulators, one 2K2 near the D669 and two 5K1. I halved each of these values so my board performed well below 40V. But look at the iraudamp7s.pdf Reference Design for a chart of resistor values vs voltage.
One A350 should work fine. Especially if you put mod decent low ESR caps into the L20D supply bypass positions. Remember the currents do surge. Also remember to connect one L20D anti-phase if you can, as the voltage pumping might upset the SMPS supply (when my A350 arrives I will check this out).
.
One A350 should work fine. Especially if you put mod decent low ESR caps into the L20D supply bypass positions. Remember the currents do surge. Also remember to connect one L20D anti-phase if you can, as the voltage pumping might upset the SMPS supply (when my A350 arrives I will check this out).
.
Last edited:
I can recommend this particular model of fan:
Newegg.com - Scythe SY1225SL12M 120mm "Slipstream" Case Fan
Needs to run at 5V for silent operation.
Honestly, I find that this fan is quiet, and adequate to keep my L20D cool. They just don't get hot
Newegg.com - EVERCOOL FAN-EC2510M12CA 25mm Case Fan
.
Hmm... 8000 RPM and less than 19 dBA?Honestly, I find that this fan is quiet, and adequate to keep my L20D cool. They just don't get hot
Newegg.com - EVERCOOL FAN-EC2510M12CA 25mm Case Fan
.
OK, I lied.. I actually use the 25mm fans from Hong Kong which are only $3.95 and have no specific experience with the NewEgg variety
The Hong Kong ones are silent.My main point was that the L20D heatsink does not get very warm unless you are testing sinewave at really high powers into a dummy load
.
Last edited:
Thanks for the info! I notice my local discount electronics emporium Rockby Electronic Components has 63v 1200uF low ESR caps on special @ $1 each at the moment - I' assume like most capacitors in PSU stages more is better here? They also have 680uF @ 80V low ESR types. I've been hunting high and low all over the place and it seems quite hard to find low ESR versions of the 100V 470uF value that the reference design uses.
re: fans: hoping my particular build, due to being undervolted runs cool enough it doesnt need them. My fanless TA3020 v3c which is my reference (+-50VDC linear PSU) only gets warm, even when caning it on a hot day.
There is not a lot of room for these capacitors on the L20D. I had to move one of the 100nF bypass capacitors under the board to make room for the capacitors I bought, which were these 330uF/100V: 8 pcs - 330uf 100v LOW ESR electrolytic capacitors | eBay
(they measured 0.05 ohms on my ESR tester).
.
(they measured 0.05 ohms on my ESR tester).
.
Last edited: