Can power amp chips be bridged in parallel?

[Mark Hammer]

I picked up a purportedly "100W" car subwoofer amp today for $13 from Princess Auto, the Canadian equivalent of Harbour Freight in the US (don't know what the equivalent might be in other countries).  The intent is to provide a simple bass amp (more about that later***) that I can throw a front end onto.  It's typical of such devices: chip is heatsinked to the chassis, wires hang off for attachment to the car's 12v battery, and there is a knobless recessed input attenuator pot.

Naturally, as soon as I got it home, I had to unscrew it and see what was under the hood.  The amp chip is a TDA7377, and is ostensibly a quad power amp chip, with 4 distinct outputs, one pair that are opposite phase to the other.  Presumably, if you expected them to provide four in-phase outputs to four separate speakers (a whopping 6W into 4 ohms each), you'd feed two of the chip inputs with a flipped-phase signal.  As is, the chip comes ready to provide two bridged-mode outputs.  The datasheet indicates that, in bridged mode, it provides two outputs of up to 20w into 4R. 

But now here's the thing.  It is a mono amp.  One input, one speaker out.  As near as I can tell, all pins relevant to use of all 4 outputs seem to be in service, going by the PCB.

So, can you use a pair of complementary amp outputs to yield a double bridged mode? If so, is there any risk entailed in doing so, or something one needs to pay attention to in order to avoid risk?  Finally, how does one calculate output power in double bridged mode?  Is it basically double whatever bridged mode delivers, or something else.

*** While certainly capable of more, the amp comes built under the assumption that it will be a subwoofer amp, most likely for those irritating people with the 15" puppies in their trunk that only play bad rap, and that only people 20ft away can hear, and not the person in the car itself.  But that's another story....what I need to do is figure out how to bypass the variable lowpass filter it includes, which the chassis legending indicates goes from 20-100hz.  If it was a thru-hole build, that would be one thing.  But apart from the power chip, a few big caps, the pots and a choke, it's SMD, on a double-sided board with the larger thru-hole components obscuring the traces, so I can't decipher which parts are associated with the filter.  The unit has no identifying marks on it or the box it came in so I can't track it down to a manufacturer.

I gather it is a fairly standard 2-pole lowpass, given the mono input/output and the use of a dual-ganged pot for the filter tuning.  The legending indicates a 20-100hz  or 5:1 tuning range.  With the pot a dual-ganged 50k, that would suggest that the pot adds onto an existing 12k or so of fixed resistance.  That, in turn, suggests I should look for a 100nf cap or two somewhere among those teeny components near the tuning pot. I say "or two" because there are several possible filter topologies; particularly given the presence of a transistor near the pot.

So that's where I'm at. 

[Gurner]

u]double[/u] bridged mode? If so, is there any risk entailed in doing so, or something one needs to pay attention to in order to avoid risk?  Finally, how does one calculate output power in double bridged mode?  Is it basically double whatever bridged mode delivers, or something else.

Not sure it's a good idea,, but IMHO it won't yield you any more output power anyway...just better spreading of the load/heat across the silicon......output power is a function of power supply @12V (being bridged the effective supply is 24V though the signal voltage won't get to that level) and your load @ 4 ohms .....neither of those change by connecting two amps in parallel. the amp might be capable of handling 100W peak (which is probably what the box was somewhar dodgily claiming), but at 12V supply and 4 ohms, it can only get you about 15W bridged.

Consider this scenario....

1 x 9V battery into 1k resistor = 9mA

2 x 9V batteries (in parallel) into 1k resistor = still 9mA

[R.G.]

Yes, power amps, including chips, can be hooked up in parallel, although it's usually good practice to insert a small value resistor in series with each output to soak up any little offset voltage differences.

Four amplifiers is a good number for a parallel-bridged amplifier; two in parallel on each side of a bridged output with the speaker strung between them. The bridging gives twice the voltage swing, the paralleling gives twice the current/power ability.

National Semiconductor did an application note on making 100W, 200W, and larger amplifiers out of the LM3886. See http://www.national.com/an/AN/AN-1192.pdf

[iccaros]

something like this?
http://www.electroniq.net/audio/tda7377-2x30-w-audio-amplifier.html

[PRR]

> "100W" ... amp ...for $13

OK, it works directly off the 14V car system, no booster supply.

One amp at 12V into 8 ohms is 2 Watts clean sine power.

Into 4 ohms, 4 Watts.

That's not enough for small car speakers and loud listening levels.

Nearly all auto-sound systems put two amplifiers per speaker in "bridge mode". This doubles the voltage, QUADruples the Watts, and also eliminates a large capacitor.

So the standard car-sound system is 16 clean Watts into each 4-ohm speaker.

(This can be rounded-up by using 14+V supply and letting THD soar to 10%; numbers like 20W even 24W per speaker are only mild mis-statements.)

My Honda factory radio is four 18W speaker outputs.

The TDA7377 claims 2X30W at 4 ohms?? Ah... read the *** footnote: "Saturated square wave output". Indeed 16W sine can be 32 Watts of Square wave. So it is an ordinary amplifier with stunningly unusual test conditions.

The TDA7377 is four amplifiers. Yes, I suspect they parallel two per side of the bridge. The honest rating is then 32 Watts in 2 ohms clean sine.

The old way to BS us was "peak power". For some BS artists this was similar to the "saturated square" test. 64 Watts of square power in 2 ohms.

I have seen "peak to peak power" which somehow doubled the number. 128 Watts. I do not see any electrical significance in this.

Find 2 ohm load, it would be comparable to any conventionally rated 30W-40W gitar/bass/PA amp. It won't fill a stadium.

BTW you need 12V-14V 5 Ampere power supply. This is not a wall-wart. They used to sell such things for running CB radios in the house. The heftier more-honest PC power supplies can put 5 Amps on the 12V rail, but they need a heavy dummy-load on the 5V/3V rail to be happy, and with ATX you need to fake a power-on signal.

The chip won't burst at 18V. On paper this would give more than double the power. However you never want to find 18V in a car (the battery would burst) and it was only $13 so I am sure it does not have heat-sinking for this level.

If you want "real power", you step-up to much more expensive amps. Traditionally they use a switcher-supply to step-up the 12V to +/-36V rails, then however many conventional +/-36V audio amp channels. These used to start at 120W total (2*60W/4*30W). I see Kenwood now starts at "350W": Kenwood KAC-1502S $45.... however it seems that listing uses funny-specs and the real rating is
60 watts RMS x 2 at 4 ohms
75 watts x 2 at 2 ohms
150 watts RMS x 1 bridged output at 4 ohms
.... so 150W tops.

And it will take 12V 15 Amperes of power to feed this. That's way past home-CB or PC supplies.

Busking? A 15 Amp-demand amp working medium loud pulls say 5A average. A car battery will run 10 hours poorly or 5 hours fine. A lawn-tractor battery will be good for a few hours.


Yes, some power-amps can parallel. Tube amps with matched gains often do fine. High-strung transistor amps usually fight each other to a smoke-out or shut-down. As was said, some series resistance limits the fighting. And auto-sound amps are fairly soft and highly protected. It might work. OTOH it is not mentioned in the TDA7377 data I found. And for $13 I do not think you have bought any assurance that it will work long-term under heavy abuse.

[DavenPaget]

> "100W" ... amp ...for $13

OK, it works directly off the 14V car system, no booster supply.

One amp at 12V into 8 ohms is 2 Watts clean sine power.

Into 4 ohms, 4 Watts.

That's not enough for small car speakers and loud listening levels.

Nearly all auto-sound systems put two amplifiers per speaker in "bridge mode". This doubles the voltage, QUADruples the Watts, and also eliminates a large capacitor.

So the standard car-sound system is 16 clean Watts into each 4-ohm speaker.

(This can be rounded-up by using 14+V supply and letting THD soar to 10%; numbers like 20W even 24W per speaker are only mild mis-statements.)

My Honda factory radio is four 18W speaker outputs.

The TDA7377 claims 2X30W at 4 ohms?? Ah... read the *** footnote: "Saturated square wave output". Indeed 16W sine can be 32 Watts of Square wave. So it is an ordinary amplifier with stunningly unusual test conditions.

The TDA7377 is four amplifiers. Yes, I suspect they parallel two per side of the bridge. The honest rating is then 32 Watts in 2 ohms clean sine.

The old way to BS us was "peak power". For some BS artists this was similar to the "saturated square" test. 64 Watts of square power in 2 ohms.

I have seen "peak to peak power" which somehow doubled the number. 128 Watts. I do not see any electrical significance in this.

Find 2 ohm load, it would be comparable to any conventionally rated 30W-40W gitar/bass/PA amp. It won't fill a stadium.

BTW you need 12V-14V 5 Ampere power supply. This is not a wall-wart. They used to sell such things for running CB radios in the house. The heftier more-honest PC power supplies can put 5 Amps on the 12V rail, but they need a heavy dummy-load on the 5V/3V rail to be happy, and with ATX you need to fake a power-on signal.

The chip won't burst at 18V. On paper this would give more than double the power. However you never want to find 18V in a car (the battery would burst) and it was only $13 so I am sure it does not have heat-sinking for this level.

If you want "real power", you step-up to much more expensive amps. Traditionally they use a switcher-supply to step-up the 12V to +/-36V rails, then however many conventional +/-36V audio amp channels. These used to start at 120W total (2*60W/4*30W). I see Kenwood now starts at "350W": Kenwood KAC-1502S $45.... however it seems that listing uses funny-specs and the real rating is
60 watts RMS x 2 at 4 ohms
75 watts x 2 at 2 ohms
150 watts RMS x 1 bridged output at 4 ohms
.... so 150W tops.

And it will take 12V 15 Amperes of power to feed this. That's way past home-CB or PC supplies.

Busking? A 15 Amp-demand amp working medium loud pulls say 5A average. A car battery will run 10 hours poorly or 5 hours fine. A lawn-tractor battery will be good for a few hours.


Yes, some power-amps can parallel. Tube amps with matched gains often do fine. High-strung transistor amps usually fight each other to a smoke-out or shut-down. As was said, some series resistance limits the fighting. And auto-sound amps are fairly soft and highly protected. It might work. OTOH it is not mentioned in the TDA7377 data I found. And for $13 I do not think you have bought any assurance that it will work long-term under heavy abuse.

The datasheet for 7377 claims 6W only when in 2R load , this is a weak 2R chipamp .

I have a alpine MRV-400 ( 4 channel , 4x 80W total into 2R 0.3% THD,  very underrated )

I opened the amp up and got surprised , massive chips ( larger then TO-3P/247 even ) , massive SMPS ( around 70V ) .
Mind you back in 1998 this was the best amp around , car amps ain't so bad after all .
• Multimode Capability
• Active Dividing Network
• Duo-ß Feedback Circuitry
• No Current Limiting
• S.T.A.R. Circuitry
• Input Mode Selector
• Continuously Adjustable Gain Control
• Fully Discrete, Complementary Output Circuitry
• Gold Plated RCA Input Connectors & Screw-Down Power & Speaker Terminals ( It's all gold but the power nuts ? it's kind of corroded slightly for no good reason )
• High Performance, Low Noise, Audiophile Quality Active & Passive Components
• Pulse Width Modulated (PWM) DC-DC Power Supply ( yes , it's a SMPS , but it's gone a long way from being noisy , it has a huge choke and 2 x 6800uf caps . )
• LED Status Monitor

[DavenPaget]

I picked up a purportedly "100W" car subwoofer amp today for $13 from Princess Auto, the Canadian equivalent of Harbour Freight in the US (don't know what the equivalent might be in other countries).  The intent is to provide a simple bass amp (more about that later***) that I can throw a front end onto.  It's typical of such devices: chip is heatsinked to the chassis, wires hang off for attachment to the car's 12v battery, and there is a knobless recessed input attenuator pot.

Naturally, as soon as I got it home, I had to unscrew it and see what was under the hood.  The amp chip is a TDA7377, and is ostensibly a quad power amp chip, with 4 distinct outputs, one pair that are opposite phase to the other.  Presumably, if you expected them to provide four in-phase outputs to four separate speakers (a whopping 6W into 4 ohms each), you'd feed two of the chip inputs with a flipped-phase signal.  As is, the chip comes ready to provide two bridged-mode outputs.  The datasheet indicates that, in bridged mode, it provides two outputs of up to 20w into 4R. 

But now here's the thing.  It is a mono amp.  One input, one speaker out.  As near as I can tell, all pins relevant to use of all 4 outputs seem to be in service, going by the PCB.

So, can you use a pair of complementary amp outputs to yield a double bridged mode? If so, is there any risk entailed in doing so, or something one needs to pay attention to in order to avoid risk?  Finally, how does one calculate output power in double bridged mode?  Is it basically double whatever bridged mode delivers, or something else.

*** While certainly capable of more, the amp comes built under the assumption that it will be a subwoofer amp, most likely for those irritating people with the 15" puppies in their trunk that only play bad rap, and that only people 20ft away can hear, and not the person in the car itself.  But that's another story....what I need to do is figure out how to bypass the variable lowpass filter it includes, which the chassis legending indicates goes from 20-100hz.  If it was a thru-hole build, that would be one thing.  But apart from the power chip, a few big caps, the pots and a choke, it's SMD, on a double-sided board with the larger thru-hole components obscuring the traces, so I can't decipher which parts are associated with the filter.  The unit has no identifying marks on it or the box it came in so I can't track it down to a manufacturer.

I gather it is a fairly standard 2-pole lowpass, given the mono input/output and the use of a dual-ganged pot for the filter tuning.  The legending indicates a 20-100hz  or 5:1 tuning range.  With the pot a dual-ganged 50k, that would suggest that the pot adds onto an existing 12k or so of fixed resistance.  That, in turn, suggests I should look for a 100nf cap or two somewhere among those teeny components near the tuning pot. I say "or two" because there are several possible filter topologies; particularly given the presence of a transistor near the pot.

So that's where I'm at. 

Still some sort of good value anyway . Why not take only the power amp circuit

[Mark Hammer]

something like this?
http://www.electroniq.net/audio/tda7377-2x30-w-audio-amplifier.html

The very same.  The data sheet does not articulate, but I imagine the "100W" claim may be related to running it on something higher than 12-14V, and slacking off on the signal quality aspects.

PRR,

The intent is certainly not to make a bass amp that would handle concert duties.  I have a modest 4 x 8" cab that I think I'll try and close the back of and port the front, and it will feed that.  A 2R load with an 18V supply should get me something that feels modestly substantial, and might also accommodate a keyboard or bass.

[R.G.]

The data sheet does not articulate, but I imagine the "100W" claim may be related to running it on something higher than 12-14V, and slacking off on the signal quality aspects.

They probably don't specify how many milliseconds it can do that for, either.    That's another favorite trick of the amp advertisers.

[Mark Hammer]

I'm regularly reminded of some $10 plastic computer speakers I saw, listed as "360W PMP"!! and they could run off 4 C-cells too!! 

[DavenPaget]

I'm regularly reminded of some $10 plastic computer speakers I saw, listed as "360W PMP"!! and they could run off 4 C-cells too!! 

I am too regularly reminded that there are actually plastic 5.1 speakers that underrate their values
I remember those damn Aiwa's 1800W PMPO !

[DavenPaget]

something like this?
http://www.electroniq.net/audio/tda7377-2x30-w-audio-amplifier.html

The very same.  The data sheet does not articulate, but I imagine the "100W" claim may be related to running it on something higher than 12-14V, and slacking off on the signal quality aspects.

PRR,

The intent is certainly not to make a bass amp that would handle concert duties.  I have a modest 4 x 8" cab that I think I'll try and close the back of and port the front, and it will feed that.  A 2R load with an 18V supply should get me something that feels modestly substantial, and might also accommodate a keyboard or bass.

Looks like it can do 40W in 2R @ 10% THD . SCARY .
I don't know what EIAJ means but i'm guessing 30% THD   Who needs overdrive when you get 30% THD  

[R.G.]

I'm regularly reminded of some $10 plastic computer speakers I saw, listed as "360W PMP"!! and they could run off 4 C-cells too!! 

Yep. They're rated at the power level where their gaseous remnants glow a bright yellow as they expand. 

[DavenPaget]

I'm regularly reminded of some $10 plastic computer speakers I saw, listed as "360W PMP"!! and they could run off 4 C-cells too!! 

Yep. They're rated at the power level where their gaseous remnants glow a bright yellow as they expand. 

They should write something below " Remnants Sold Separately , Not Available for sale in Canada "
The remnants might glow orange depending on mood of the bullshit-slapper

[PRR]

> for those irritating people with the 15" puppies in their trunk that only play bad rap, and that only people 20ft away can hear

It isn't really good for that. 30W-40W in an auto-sound subwoofer will hardly rattle your luggage. It is scarcely more than my 36W in two 6x9 deck speakers, little improvement over the good/better factory system.

The annoying guys are running hundreds of honest Watts (thousand-watt on the marketing BS). That's why I had the 4*40W Kenwood... to be only mildly annoying. But one 40W is just wannabe boom-car. Or maybe emergency-repair limp-home.

This is all partly because auto-sound subwoofers have to be very low efficiency to reach deep notes in small spaces. If you allocate SPACE, like 3/4 of an Econoline, then you can run more efficient speakers. Four JBL 2220/E150 in the dividing wall could give very strong bass with just 40W. Obviously the cost and size make lamer speakers and bigger amps the way-to-go.

> how to bypass the variable lowpass filter



Put ohm meter on pin 4/5. Find the cap which is 0.0 ohms to pin 4/5. Bust it off. Same for pin 11/12. (There may be just one cap for both inputs... at $13 this seems likely.) Being SMD you will probaby poke any visible cap-tab with your smallest iron until it comes loose. Damage is not a problem.

Get small wire. Tin it a few places. Tack it to pin 11/12, then to pin 4/5, then to a 1uFd cap (the "0.1u" on my sketch is a mistake). (The jumpers 4-5-11-12 may already be in there, if so then just tack whichever pin is easy.)

Call the other end of this cap "input", with ~~10K impedance and -3db @ 18Hz. (0.33u-0.22u 60Hz-80Hz may be more guitar-friendly or physically smaller.)

If electrolytic, "+" side of cap goes toward chip.

------

> gaseous remnants glow a bright yellow

Exciting but unlikely. It will shut-down for many things. Car wiring is full of accidents (ask my truck), car-sound makers hate warranty returns, the chip-makers try to cover all accidents.

> datasheet for 7377 claims 6W only when in 2R load

I see 6W in 4r or 10W in 2r.... maybe different copies have different typos?
http://www.e-ele.net/DataSheet/TDA7377.pdf

> what EIAJ means

EIAJ is/was a manufacturer-friendly standards group. The intent is limited-BS; but the standards can be interpreted in BS ways.

BTW, "Saturated square wave output" is also a low-dissipation condition. It might run that way for minutes or years, yet shut-down swiftly on Sine waveform. "Normal" speech/music signals can be less stressful than Sine. Overdriven guitar can be hard work.

BTW: "Figure 20: The New Output Stage".... this scheme was used in Dan Meyer's "Universal Tiger" amplifiers in 1970, for much the same reason, and was not new then.

[DavenPaget]

> for those irritating people with the 15" puppies in their trunk that only play bad rap, and that only people 20ft away can hear

It isn't really good for that. 30W-40W in an auto-sound subwoofer will hardly rattle your luggage. It is scarcely more than my 36W in two 6x9 deck speakers, little improvement over the good/better factory system.

The annoying guys are running hundreds of honest Watts (thousand-watt on the marketing BS). That's why I had the 4*40W Kenwood... to be only mildly annoying. But one 40W is just wannabe boom-car. Or maybe emergency-repair limp-home.

This is all partly because auto-sound subwoofers have to be very low efficiency to reach deep notes in small spaces. If you allocate SPACE, like 3/4 of an Econoline, then you can run more efficient speakers. Four JBL 2220/E150 in the dividing wall could give very strong bass with just 40W. Obviously the cost and size make lamer speakers and bigger amps the way-to-go.

> how to bypass the variable lowpass filter



Put ohm meter on pin 4/5. Find the cap which is 0.0 ohms to pin 4/5. Bust it off. Same for pin 11/12. (There may be just one cap for both inputs... at $13 this seems likely.) Being SMD you will probaby poke any visible cap-tab with your smallest iron until it comes loose. Damage is not a problem.

Get small wire. Tin it a few places. Tack it to pin 11/12, then to pin 4/5, then to a 1uFd cap (the "0.1u" on my sketch is a mistake). (The jumpers 4-5-11-12 may already be in there, if so then just tack whichever pin is easy.)

Call the other end of this cap "input", with ~~10K impedance and -3db @ 18Hz. (0.33u-0.22u 60Hz-80Hz may be more guitar-friendly or physically smaller.)

If electrolytic, "+" side of cap goes toward chip.

------

> gaseous remnants glow a bright yellow

Exciting but unlikely. It will shut-down for many things. Car wiring is full of accidents (ask my truck), car-sound makers hate warranty returns, the chip-makers try to cover all accidents.

> datasheet for 7377 claims 6W only when in 2R load

I see 6W in 4r or 10W in 2r.... maybe different copies have different typos?
http://www.e-ele.net/DataSheet/TDA7377.pdf

> what EIAJ means

EIAJ is/was a manufacturer-friendly standards group. The intent is limited-BS; but the standards can be interpreted in BS ways.

BTW, "Saturated square wave output" is also a low-dissipation condition. It might run that way for minutes or years, yet shut-down swiftly on Sine waveform. "Normal" speech/music signals can be less stressful than Sine. Overdriven guitar can be hard work.

BTW: "Figure 20: The New Output Stage".... this scheme was used in Dan Meyer's "Universal Tiger" amplifiers in 1970, for much the same reason, and was not new then.

They don't just do a few hundred watts , most i seen do a few thousand watts on JL's . Rich assholes who know nothing about proper bass .
I have 2 subs and that's all classic subs , JBL GT82D only 100Wrms but at 93db/W who argues ? ( Sounds very epic when used for my guitar amp )


I was kinda drunk when i wrote the weak chipamp comment ... Oh well it is still weak anyway .

[brett]

Hi
are any of these things Class D? ie can run above the supply for a while
I assume that D class have little charge pumps or similar, but someone else will know what's going on.

A few years ago I bought a 60W car amp for $20 (a Sony?) from a pawn shop and was impressed to see that it had little toroids in it and the power rail ran quite high (60V ?). I'm guessing it was a $200 amp new.  It looked like the power supply used MOSFETs, which made sense (for reasons I can't remember now).
It would rattle my shed and flatten a gel motorcycle battery (~7Ah) after about 2 hours.
cheers

[PRR]

> are any of these things Class D? ie can run above the supply for a while

"run above the supply for a while"?

No, none of the amps mentioned here are Class D (except perhaps Daven's spiffy MRV-400).

Class D does not directly address "supply" issues. 12V supply in 4 ohms is 4 Watts Sine, whether A, B, or D.

In A and B, the two output transistors work as "faders", letting current to the speaker and wasting the excess voltage, smoothly.

In class D the output transistors are switches, Just ON and OFF. But on/off so rapidly, in specific ratio, that it has the effect of smooth fading.

That is not simple to do. (The earliest proposals were 1960s or earlier, but practical class D didn't happen until the 1990s for subwoofers and the 2000s for full-range audio.)

Once an amp gets THAT fancy, it may be expedient to also build-in supply boosters.

> I bought a 60W car amp ... it had little toroids in it and the power rail ran quite high (60V?). ... It looked like the power supply used MOSFETs, which made sense

Yes, the classic BIG car-sound amp from 1970s until today brings the 12V to a DC/DC converter. Same as a PC power supply. The PC supply converts 320VDC (doubled from 120V AC) to 5V/12V; but simple re-design can convert 12V to +/-40V or almost anything desired. Now the audio stage is basically the same as the home hi-fi. Yes, MOSFETs are common, especially since car-sound amps can be too-small for their power and can run HOT, and at extreme temps the MOSFETs can be easier to manage.

> It would rattle my shed

My Kenwood was an honest 35+Watts each of four channels, and intended to run "bridged" as two 140W outputs. That is a LOT of power.

flatten a gel motorcycle battery (~7Ah) after about 2 hours.

12V*7AH is 84 Watt-Hours, or average demand 40 watts, which implies maybe 20 Watts to the speaker *average*, which might be 60++ Watts speech/music at or past clipping.

[DavenPaget]

> are any of these things Class D? ie can run above the supply for a while

"run above the supply for a while"?

No, none of the amps mentioned here are Class D (except perhaps Daven's spiffy MRV-400).

Class D does not directly address "supply" issues. 12V supply in 4 ohms is 4 Watts Sine, whether A, B, or D.

In A and B, the two output transistors work as "faders", letting current to the speaker and wasting the excess voltage, smoothly.

In class D the output transistors are switches, Just ON and OFF. But on/off so rapidly, in specific ratio, that it has the effect of smooth fading.

That is not simple to do. (The earliest proposals were 1960s or earlier, but practical class D didn't happen until the 1990s for subwoofers and the 2000s for full-range audio.)

Once an amp gets THAT fancy, it may be expedient to also build-in supply boosters.

> I bought a 60W car amp ... it had little toroids in it and the power rail ran quite high (60V?). ... It looked like the power supply used MOSFETs, which made sense

Yes, the classic BIG car-sound amp from 1970s until today brings the 12V to a DC/DC converter. Same as a PC power supply. The PC supply converts 320VDC (doubled from 120V AC) to 5V/12V; but simple re-design can convert 12V to +/-40V or almost anything desired. Now the audio stage is basically the same as the home hi-fi. Yes, MOSFETs are common, especially since car-sound amps can be too-small for their power and can run HOT, and at extreme temps the MOSFETs can be easier to manage.

> It would rattle my shed

My Kenwood was an honest 35+Watts each of four channels, and intended to run "bridged" as two 140W outputs. That is a LOT of power.

flatten a gel motorcycle battery (~7Ah) after about 2 hours.

12V*7AH is 84 Watt-Hours, or average demand 40 watts, which implies maybe 20 Watts to the speaker *average*, which might be 60++ Watts speech/music at or past clipping.

My MRV-400 is a Class B .

[betula]

Hi there,
First off....I want to thank you for creating this forum. Yes, I'm a newbie.
This is more of a question than a response.
I purchased one of those subwoofer amps at Princess Auto.
I wondering about the hookup to power, input and speaker.
The red wire(with fuse) is obviously power. The black is Ground, most likely.
However, the blue wire I'm  not sure off. I opened the box and notices that there
was small print saying "Rem". Can anyone tell me what the blue wire is for and where
it should get connected to? Right now I can't get the amp to work.
Thanks,
Betula