DOD 650 mini-amplifier

[PRR]

> need to listen to it, or mic, in some very special ways.

A 4*12 has 64 times the cone-area of one 3". Air is mushy. 64 times the area is much better grip on the air. It is not unlikely that, in mid-bass, the 4*12 with 0.6 Watts will be as loud as the 3" with 38 Watts (if it survived).

No, a flea-Watt amp into a Full Stack will not fill a stadium! But we did fill 3,000-seat stadiums (no PA) with 100-200W in full stacks. It seems reasonable that a flea into a stack could be Real Frikkin Loud in a 15-seat venue (large living room or teeny club). Without tricks.

[Mark Hammer]

Normally, I would humbly defer to you, because of your generally superior electronic knowledge andgreater mathematical dexterity.  But here I will draw attention to the relative mass of the speaker cones, and plain old vanilla inertia

My wife got us a new kitchen sink last year with "hands-free" operation.  Trouble is, the thing runs off batteries and those damn solenoids eat a LOT of current to operate.  So we just disconnected the automatic stuff and use it manually because we got tired of crawling under the counter to change the batteries.

Speakers are essentially solenoids, and it takes juice to move those bloody cones.  The bigger the voice coil and cone mass, the more current.  And if they are intended to survive the sort of current accompanying 100 or even 50W, they are not going to be particularly light or compliant.  The puny 600mw of a 386-based amplifier will certainly get FAR more bass from those speakers than from a 3" speaker in a cab 1/500th the internal volume (although in many instances the mini-amp will have a insufficiently large electro cap on the output to permit much below 200hz).  BUt that's going to be more a result of cab/speaker resonance, and the inability of the mini-amp to make the speaker move fast enough to reproduce treble.

Of course, when all is said and done, what the battery-operated mini-amp can or can't make the speaker do is completely separate from how the combo sounds.  Indeed, it may well be the case that what some folks find appealing about driving a cab meant for much greater wattage is a direct result of what the mini-amp can't do.  Heck, that principle is a big part of what rock music is supposed to sound like.  Overdrive and distortion is a product of a signal path that simply can't amplify with any degree of fidelity.

[PRR]

> those damn solenoids eat a LOT of current

Bad design?

1) I wonder if they take power to stay "open". Simple but wrong. A shuttle valve only needs a short blip to open or close, needs no power to hold its state. Like normal relays versus latching relays.

2) For a fraction of what they charged you they cudda put a mini-turbine in the water path to spin a 59-cent Mabushi motor as a generator and recharge batteries.

They are not optimizing the motor to the load and the cost of power. Such concerns are a *prime* obsession with the speaker engineer.

The ideal speaker is like many other engineering devices: 90+% efficiency. Motors and transformers and gearboxes do this routinely. Combustion engines are awkward because they don't beat 40% efficiency.

The "ideal" speaker for highish efficiency down to 50Hz must have a 120 inch (10 foot, 3m) air-smacker. If any smaller, it can't couple to air well over a useful wife freq band.

Near-Ideal speakers are impractical. (Theater speakers of the 1930s came close; so do some of the very large arrays now used in cow-palaces.)

The 3-inch cone is efficient down to 2,000Hz. Below that efficiency % drops as square of frequency. At 200Hz it is necessarily 1% efficient at best. For "flat" response (and a mass-related reason) the parameters are adjusted to be 1% efficient 200Hz-2KHz. (Above that, efficiency falls but directivity rises; on-axis response may be flat another few octaves up.)

Take a four-12 array as a 24" cone. 8 times bigger than the 3-inch. All frequency benchmarks drop 8X. It loses efficiency below 250Hz (not 2KHz). For other reasons it will never come close to 100%, even 50%, efficient. For guitar we would likely aim for 80Hz, 3X lower freq than the bass/efficiency corner, 10X lower efficiency, around 10% from 80Hz to 250Hz. This size flat array also leads to high directivity in the >250Hz range, so stunning "throw", impressive sound in-front which cuts-through reverberant spaces. Above 1KHz-2KHz the directivity pattern gets very narrow while also growing side-lobes different at every frequency, though a four-12 will never keep all four cones-phase-matched above 500Hz which mixes-up the otherwise beamy directivity.

Unless heavily slugged for deep bass, or assembled by faucet designers, the larger air-slapper will generally put out more sound per electric Watt.

You seem to be leaning on an argument that the amp power and speaker rating must be alike. The ideal speaker is perfectly linear at any power. Yes, MUCH guitar tone is from driving non-ideal non-linear speakers somewhat past what the original (non-guitar) designer ever expected. But some speakers got past the Magnavox designs. The older Altec theater woofers, E-V's EVM series, and the wonderous JBL D-120/130. Some players dislike them as "too clean". Others appreciate clean reproduction of whatever mangled tone you feed them. Running 0.6 Watts into four Brand E speakers rated 25W/each with heavy breakup is a lot like a 50W into an EVM good for 600+W clean peaks.

[Mark Hammer]

You seem to be leaning on an argument that the amp power and speaker rating must be alike. The ideal speaker is perfectly linear at any power. Yes, MUCH guitar tone is from driving non-ideal non-linear speakers somewhat past what the original (non-guitar) designer ever expected. But some speakers got past the Magnavox designs. The older Altec theater woofers, E-V's EVM series, and the wonderous JBL D-120/130. Some players dislike them as "too clean". Others appreciate clean reproduction of whatever mangled tone you feed them. Running 0.6 Watts into four Brand E speakers rated 25W/each with heavy breakup is a lot like a 50W into an EVM good for 600+W clean peaks.

I guess it may seem like that, but that's not my intent at all.  The gist of my argument is that few speakers are "ideal".  They will have a linear and a non-linear range, plus some transitional zone.  Some minimum given amount of amplifier output current/voltage is required to nudge them into their linear range, and the bigger and heavier the cone-and-coil assembly, and the stiffer the spider (in anticipation of handling BIG wattages), the more juice will be needed to move from non-linear to linear.  It doesn't have to be a LOT - which is why single-ended three-tube 5W amps driving a 4x12 cab can sound pretty darn fine - but nominal 1W amps running off a 9V battery are highly unlikely to meet those minimum requirements.

Again, what transpires when a teeny battery amp tries to push a big cab may well sound great; especially when mic'd up just right.  But the tone achieved is unlikely to be the result of any part of that signal chain operating as nature and design intended.  In other words, it's not the result of big speakers adding "more" to the mini-amp's output than the internal small speaker, the way that replacing a 10 with a 12 in a medium-powered (under 25W) amp would.

[PRR]

> few speakers are "ideal".

None.

> They will have a linear and a non-linear range, plus some transitional zone.  Some minimum given amount of amplifier output current/voltage is required to nudge them into their linear range

Ah.

The "center" of a voice coil's motion should be very darn linear. They should reproduce the smallest quiver of wide dynamic range material. No stiction.

If the linear range did not extend *through* zero, that would amount to "crossover distortion" as in transistor amplifiers. Nasty at low level.

My experience with speakers says they make no non-linear crap at levels at least 70dB down from "normal listening level". If they did, much "fine" music would sound crappy in the soft passages. I suspect any near-zero stiction happens at levels well below room ambient background noise.

[Frances Rhodes]

hi everyone

bumping this old thread.

i recently build a clone of the DOD 650 using that schematic, and it works, bad, but it works.
i build it using salvaged speakers from an old TV set and they are rated 10W each so i was looking for a way to increase power without having te redesign a new power amp section, and i looked at the schematic again and something bothers me about the 2 pairs of output transistors.
isn't the lower darlington pair strange, with the emitter of the first npn pointing "up"? or is it one of the different ways to hook 2 transistors up for more gain?
the only way of making a darlington pair i know of (but i know few...) is to tie the second base to the first emitter and tie both collectors together, so this setup doesn't look right to me. since my amp is making sound, means it works, but doesn't mean it works the way it's supposed to, so i may be wrong but i feel like this odd-looking transistor arrangement may be a reason why the sound bad.

cheers

[PRR]

Not Darlington. Sziklai.



> it works, but doesn't mean it works the way it's supposed to

No clue. I can't see your voltages from here.

[Frances Rhodes]

i'm not talking about the sziklai pair (PNP + NPN) but the other one, the lower one (NPN + NPN) that looks like a darlington but with the first transistor upside down, with the second transistor tied to its collector instead of its emitter.

i can post voltages, but my question is about that schematic, for now.

[PRR]

I see. That's just wrong. Where did this plan come from?

Should be this.

[PRR]

me> Where did this plan come from?

http://music-electronics-forum.com/t12696/#post100112
Mark Hammer: "....here is the schematic. FWIW, I'm not so sure it is entirely accurate, even though it came directly from DOD.:"

[PRR]

> ...a clone of the DOD 650 .... salvaged speakers from an old TV set and they are rated 10W each so i was looking for a way to increase power....

9V supply into the 8+8= 16 Ohm load is only 1/2 Watt. Probably 1/4 Watt due to opamp and output losses.

The indicated 2N4124/5 transistors are very low-current jobs, already strained at this power level.

Your "10 Watt" speakers won't stand 10 Watts of distorted guitar for very long.

I do wonder if your TV speakers are 8 or 4 Ohm.

Short path to "more power": get a Car-Audio chip. Run on 9V it will make about 9 Watts in 4 Ohms (two 8r parallel), which they will probably stand. The 9V supply needs about 800mA, so a fairly hefty supply.

The preamp from the 650 will be useful.

[Frances Rhodes]

ok so you think that the problem here is that the NPN should be a PNP instead, that makes sense.

my salvaged speakers are 10W / 8 ohms each, in series to match the 16 ohms output specified.

i was thinking of keeping the same design (considering the modification you made) replacing the four output transistors by two TIP31 and two TIP32 which can handle more power, and also use a "bigger" supply, maybe 12V-1A. can this work or is it a completely wrong way of doing it?

[PRR]

12V DC allows 4Vrms swing in a load. That is 2 Watts in a 8 Ohm load or 4W in two 8r loads parallel. Naked TIP devices will just barely support the dissipation. Any short will burn the TIPs.

[Frances Rhodes]

"naked" you mean without a heat sink?

[bluebunny]

Yep.

[Frances Rhodes]

ok, right, but other than that, with a proper supply, this should work fine with just swapping the actual transistors for TIP ones?