Threads merged.
Thank you, kind sir.
Hank, I didn't check your math. It looks right.
One amp's output simply alternates above and below a reference. In bridge mode, the other amp's output does the same, but with inverted polarity. So there is twice the voltage at the load, and the +/- reference becomes somewhat "virtual", for lack of a better term. It's no longer ground or the midpoint of either amp individually.
Hank, I didn't check your math. It looks right.
One amp's output simply alternates above and below a reference. In bridge mode, the other amp's output does the same, but with inverted polarity. So there is twice the voltage at the load, and the +/- reference becomes somewhat "virtual", for lack of a better term. It's no longer ground or the midpoint of either amp individually.
I saw it, but I'm not understanding why not, as one leg traverses up, the other leg traverses down....I'll do some more scoping...in the words of Arnie, I'll be back....
Ok, just revisited this...
yellow trace is the amp's +ve leg
red trace is the amp's -ve leg
purple trace is the 'difference' between the two (USB scope setting A-B), so it looks to me now as if the voltage peak to peak is doubled...
(yes, I know there's ringng/noise at the peaks, but it was just a quick lash up!)
So for the above, 7.22V(-ish) is peak, therefore 5.105V RMS ...coil resistance is 8 ohms (again for simplicity, not taking into account its reactance!) - therefore 5.1505/8 = 0.644A through the speaker coil??
yellow trace is the amp's +ve leg
red trace is the amp's -ve leg
purple trace is the 'difference' between the two (USB scope setting A-B), so it looks to me now as if the voltage peak to peak is doubled...
An externally hosted image should be here but it was not working when we last tested it.
(yes, I know there's ringng/noise at the peaks, but it was just a quick lash up!)
So for the above, 7.22V(-ish) is peak, therefore 5.105V RMS ...coil resistance is 8 ohms (again for simplicity, not taking into account its reactance!) - therefore 5.1505/8 = 0.644A through the speaker coil??
Last edited:
completely untrue.
A 4ohm capable amplifier will not be capable of driving a 2ohm speaker.
In a similar vein, when you bridge the two 4ohm capable amplifiers and connect them to a 4r0 load you will not get four times the output power.
The rule is as I stated:
No matter how many times you see "4times the power" it is still wrong.
My 2cents.
Just confirmed the above caclulation with a real world reading by putting my DVM in series with the power supply & reading the DC current drawn by the overall circuit ...it all tallies (give or take a couple of mA). So for the purposes of calculating current through the coil, whatever peak to peak voltage is scoped on one leg of the BTL amp is Vpeak.
Last edited:
It may well be that there's no complicated mental dexterity needed ...but the differing/conflicting info in some of the replies suggests it's not done & dusted in everyone's else's heads either!
It actually complicates it (to my eyes at least) as considering this a BTL as one amp first ....because it's not one amp, but two amps operating 180 degrees out of phase...which is why I wanted to square this away in BTL terms (vs SE)...which I've just done by confirming the 'on paper' calculations with actual real world outcome - I'm happy with this now
Last edited:
each amp does "see" 1/2 of the attached load, with the opposite end of the load going the other way the midpoint of the load is a virtual gnd
if each amp in the bridge can drive the 2x current into the 1/2 load Z it "feels" on its output then you do get 4x power into the same load Z
having current reserve to double current for load dips to 1/2 the nominal load is the bare minimum reserve for a decent audio power amp given the demands of some loudspeakers (worse with some crossovers)
so if you want the same current drive reserve, say reasoning form a audio power chip amp's datasheet bullet item "8 Ohm" power rating then you probably want to parallel 2x of each chip in the bridge for a total of 4 amp chips to give 4x the power into the same nominal impedance load
if each amp in the bridge can drive the 2x current into the 1/2 load Z it "feels" on its output then you do get 4x power into the same load Z
having current reserve to double current for load dips to 1/2 the nominal load is the bare minimum reserve for a decent audio power amp given the demands of some loudspeakers (worse with some crossovers)
so if you want the same current drive reserve, say reasoning form a audio power chip amp's datasheet bullet item "8 Ohm" power rating then you probably want to parallel 2x of each chip in the bridge for a total of 4 amp chips to give 4x the power into the same nominal impedance load
Hi,
Its not completely untrue. For car audio a single simple amp can put out about
5W into 4 ohms and a pair of simple BTL amplifiers about 20W into 4 ohms.
Of course this implies each amplifier of the BTL pair must be 2 ohm capable.
If they are then "4 times the power is true".
Of course if each amplifier is only rated down to 4 ohms and has a fixed
current limit related to 4 ohms loading, bridging the amplifiers will give
you no more power into 4 ohms, but will give you twice the power into
8 ohms, and 4 times the power into 16ohms.
What you can say is always true but not always applicable is :
"Bridging into twice the load impedance always gives you twice the power."
Bridging into the same load impedance can give you
anything between no extra power and 4 times the power.
For HiFi going bridged is usually applicable for 8 ohm speakers and 4 ohm
capable amplifiers. If the stereo amp is say 100W 8 ohm, 150W 4 ohm,
then bridging into the same same speaker will give you 3 times the power.
rgds, sreten.
Last edited:
No, it is not true.
If you to take extreme car amplifiers as an example, then look at the real specifications.
I take a hypothetical example that can drive 1r0 test load.
max Vac . . load
. 20V . . . 1r0
. 22V . . . 2r0
. 23.5V . . 4r0
. 23.9V . . 8r0
This amplifier is rated to drive a 2ohm speaker load.
Into 2ohm it will deliver a maximum of 22^2 / 2 = 242W into 2r0
into 4r0, 138W, into 8r0, 71W.
Now let's bridge these amplifiers.
into 2r0, 800W, but this bridged amp is not capable of driving a 2ohm speaker.
into 4r0, 484W, into 8r0, 276W, into 16r0, 142W.
There is no 4times for any of the test loads.
If you to take extreme car amplifiers as an example, then look at the real specifications.
I take a hypothetical example that can drive 1r0 test load.
max Vac . . load
. 20V . . . 1r0
. 22V . . . 2r0
. 23.5V . . 4r0
. 23.9V . . 8r0
This amplifier is rated to drive a 2ohm speaker load.
Into 2ohm it will deliver a maximum of 22^2 / 2 = 242W into 2r0
into 4r0, 138W, into 8r0, 71W.
Now let's bridge these amplifiers.
into 2r0, 800W, but this bridged amp is not capable of driving a 2ohm speaker.
into 4r0, 484W, into 8r0, 276W, into 16r0, 142W.
There is no 4times for any of the test loads.
hi,
Your just being tedious about the simple fact an amplifier never usually
delivers double the power into a half impedance load, and we all know
that is true, irrespective of bridging, 4 ohm power is not 2 times 8 ohm.
Say something useful, rather than negating the impossible.
rgds, sreten.
It is stupid to say an amplifier capable of driving a 1R load cannot
drive 2R loads bridged. its also stupid to simply say a 8R rated
amplifier can't drive 4R loads, of course it can up to half power.
Last edited:
If the test load Andrew is referencing is a 1R0 resistor, then the amp will blow up driving a 1R0 reactive speaker load. Speakers are not resistors, and the output transistors will feel the difference.
Overloading an amp but only turning it up half way does stress the amp more, it's not a 50-50 ratio because load Z is not closely matched to the Zout. Besides, this is bad practice because at some time some wank-er will crank the volume knob and 'puff' will go the output stage.
Overloading an amp but only turning it up half way does stress the amp more, it's not a 50-50 ratio because load Z is not closely matched to the Zout. Besides, this is bad practice because at some time some wank-er will crank the volume knob and 'puff' will go the output stage.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.