Is their a basic equation or calculator for figuring power output of an un-bridged transistor AB amplifier based on the voltage of the power supply?
I'd like to know what +/-32v into 8 ohms will do just before clipping. How about into 4 ohms? How about +/-42v into 8 ohms? etc etc...
I have to rewind some boost supplies to do 100w into 8 ohms. Trying to find quick math that'll get me in the ballpark.
Happy New Year everyone!
I'd like to know what +/-32v into 8 ohms will do just before clipping. How about into 4 ohms? How about +/-42v into 8 ohms? etc etc...
I have to rewind some boost supplies to do 100w into 8 ohms. Trying to find quick math that'll get me in the ballpark.
Happy New Year everyone!
Yes, the theoretical maximum power output with a +/- V supply is: V^2 / ( 2 x Rload )
The peak output voltage is V; the Vrms output is ( V / sqrt 2 ); and the output power is: ( Vrms^2 / Rload )
Then you have to consider: power supply regulation (drop under load), conduction drop of output devices,
driver stage limitations, etc.
Last edited:
Do I use the total rail to rail voltage?
I used 32v below.
V^2 / ( 2 x Rload ) gets me 64w into 8 ohms. That sound right?
( Vrms^2 / Rload ) gets me 5.64w into 8 ohms. I used 1.41 for ^2
nvm I'm screwing something up.
For rms.
Say 32-0-32 supply.
32/1.41=22.7Vrms
22.7*22.7=515.3
515.3/8=64.4w
That right? Or was I supposed to use 2*Rload?
I used 32v below.
V^2 / ( 2 x Rload ) gets me 64w into 8 ohms. That sound right?
( Vrms^2 / Rload ) gets me 5.64w into 8 ohms. I used 1.41 for ^2
nvm I'm screwing something up.
For rms.
Say 32-0-32 supply.
32/1.41=22.7Vrms
22.7*22.7=515.3
515.3/8=64.4w
That right? Or was I supposed to use 2*Rload?
Last edited:
Doug Self's website shows all the calculations to figure it out from raw AC from the tranformer.
The Signal Transfer Company: Power Output
The Signal Transfer Company: Power Output
Just use the formula as above. For +/- 32V supplies, P = 32^2 / (2 X 8) = 64W.
The last way you did it was ok too, but you asked for the output power as a function of the supply voltage.
This is a lot shorter formula than what is on Self's site. It's likely as accurate too considering an 8 ohm speaker will measure a different resistance at any given frequency anyway. Wattage ratings in amplifiers are like horsepower ratings in cars.
Bear in mind that many other factors limit the maximum power output. This is just the absolute maximum possible with no other considerations.
Self's calculations go a little more in depth. He calculates Re resistor values into the equation. He also states the calculations are flawed due to changing load.
I'm just trying to get in the ballpark. I'm making a 4ch amp because the best sounding speakers I found for my car's doors happen to be 8 ohm. The rears are 4 ohm. The amp is 4 power supplies and 4 amp modules in one chassis. The two supplies for the front channels have to be rewound for a higher voltage. I'm trying to achieve 100x2@4ohm and 100x2@8ohm but I don't know what voltages I need.
This I know about and horsepower is pretty exact.
This I know about and horsepower is pretty exact.
Most amps use emiter followers for the output stage and never output the whole output voltage. Using a bypolar emiter output stage you will typicaly clip ~ 3v before you reach the rail voltage and in some amps this voltage loss may even be higher. Then your voltage supply will probably drop a volt or more under load. All in all using a 32v + & - supply you are probably looking at about 45 to 50 W rms at most.
I'm just trying to get in the ballpark. I'm making a 4ch amp because the best sounding speakers I found for my car's doors happen to be 8 ohm. The rears are 4 ohm. The amp is 4 power supplies and 4 amp modules in one chassis. The two supplies for the front channels have to be rewound for a higher voltage. I'm trying to achieve 100x2@4ohm and 100x2@8ohm but I don't know what voltages I need.
This I know about and horsepower is pretty exact.
42 volt rails should get you around 100 watts. Sounds like a neat amplifier.
Torque means much more than horsepower too.
Then I'm probably looking at rewinding for 52v at least due to the sag. I'm using an ebay SMPS rated for 180w. I had it putting out 100w continuous during a test and the input caps were getting pretty toasty.
Torque? Here we go again. lol Torque matters to those who mis-understand the function of a transmission.
Gear your car correctly and stay in the power band and horsepower is the only number you'll care about. Trust me. I own a gutless RX-8 with zero torque. If I keep the revs up it's plenty fast. I'm currently swapping in a 3.2L Isuzu v6 (don't ask) and twin turbocharging it. Google "Isuzu v6 RX-8 swap" and look at the images. Stay out of the forum though. There's 300+ pages of them burning me at the cross for doing it. Your AC voltage from the transformer will be lower than your DC voltage at the rails. I made the horsepower comparison because it's equally not an ideal way to rate something. The VA of the power supply is what does the work in an amp. You can raise the voltage and get a more impressive power rating but's really gaining nothing without raising the current available along with it. Just like your car engine. You can spin it faster and get a higher horsepower number but it doesn't gain you anything. Crank up the boost a bit and add some more torque and the fun begins.
I hope your kidding. Or I'm not following you. That's like saying amps are all that matter and watts won't gain you anything.
Like horsepower, watts are what matter. That's the bottom line. I'd rather produce those watts using voltage, especially since I'm limited by a high impedance load. The car audio industry has it the other way around. They like to have the current do the work by driving very low ohm loads. I think they do that for safety reasons. Imagine the voltage output to do 1kw into 8 ohms.
Would you rather have 2000ft of torque at 550 RPM? Or 200ft at 5500 RPM? They are both 210hp.
There's nothing wrong with going higher voltage in an amplifier. I'm just saying you need to boost current along with it. They go together.
Same with your engine torque and horsepower go together. I personally would love 2000 lbs of torque at 550 rpm if there was a transmission that could handle it at that low an RPM. Heavy truck transmissions are still pounding holes in gear faces at 1200 rpm though. I have better things to do than shift gears all day. I'd rather just whack the throttle and go. This is 30 years of custom vehicle building experience talking. All my customers feel the same. They've learned the hard way as well.
Why do I have to increase current? Say I have 4a at 32v. That's 128w. Now if I rewind it for 64v and reduce the current to 2a I still have 128w. I'm not asking the supply to work any harder then it was. I may even gain a little power since the losses would be less at lower current.
If you lined up a 100hp Harley Davidson (low rpm, high torque) against a 100hp sportbike (high rpm low torque), who wins?
Don't forget there is a load attached to the amp. When you increase voltage across a load, the current through it increases. This is why you always have to uprate the current for higher rails, not doing so will allow the supply to sag and reduce power output (to the point that it would've made more sense to use lower rails).
35V swing allows 100W output, so if you use bipolar devices 42-45V rails work fine and give you some headroom if the supply is stiff enough. Normally the power supply should be able to deliver at least twice the current required to reach maximum output (found in exactly the same way as the first set of equations) for real-world speakers, which are reactive. Some would say more. So yes, a higher voltage power supply needs to provide more current else it is pointless.
1kW into 8 ohms is easily generated with approximately 100V rails, and is par for the course for most professional amplifiers. And most can do it all day. The reason that cars do it using current is to minimise switching losses from 12V batteries, as the higher the output voltage, the lower the efficiency. I have heard of some competition vehicles using batteries in series to help out with this particular issue.
35V swing allows 100W output, so if you use bipolar devices 42-45V rails work fine and give you some headroom if the supply is stiff enough. Normally the power supply should be able to deliver at least twice the current required to reach maximum output (found in exactly the same way as the first set of equations) for real-world speakers, which are reactive. Some would say more. So yes, a higher voltage power supply needs to provide more current else it is pointless.
1kW into 8 ohms is easily generated with approximately 100V rails, and is par for the course for most professional amplifiers. And most can do it all day. The reason that cars do it using current is to minimise switching losses from 12V batteries, as the higher the output voltage, the lower the efficiency. I have heard of some competition vehicles using batteries in series to help out with this particular issue.
You're talking about racing a Harley so this is obviously going nowhere. Go ahead and try it out. Hopefully it works out for you.
On a side note, here's a "Harley" that might change your outlook on lower rpm engines but there's no actual "Harley" parts in it. 1000 HP, all in by 5500 rpm. No transmission. Dump the clutch and hang on.
On a side note, here's a "Harley" that might change your outlook on lower rpm engines but there's no actual "Harley" parts in it. 1000 HP, all in by 5500 rpm. No transmission. Dump the clutch and hang on.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.