"Myths"

[Mark_ab 5,000+ posts]

Boo-yeah! hah

Arguing with Josh = //content.invisioncic.com/y282845/emoticons/nono.gif.eca61d170185779e0921b0faa9704973.gif

 

[more_spl 10+ year member]

//content.invisioncic.com/y282845/emoticons/verymad.gif.3f39c5c2fd57527b671fad3efdfac756.gif

wtf is this distortion doesnt kill speakers !!!

thast what we've all been saying, even the smarter regulars on this board.

what's with this distortion doesnt kill speakers all of a sudden?

isn't it like this?

Too much power= push speaker past limits= cause of distortion= blown speakers?

 

[jlaine 10+ year member]

Originally posted by more_spl //content.invisioncic.com/y282845/emoticons/verymad.gif.3f39c5c2fd57527b671fad3efdfac756.gif

wtf is this distortion doesnt kill speakers !!!

thast what we've all been saying, even the smarter regulars on this board.

what's with this distortion doesnt kill speakers all of a sudden?

isn't it like this?

Too much power= push speaker past limits= cause of distortion= blown speakers?

Incorrect..

The only even remotely plausible way to blame distortion is via the amplifier, and the phenomenon known as clipping. Pushing a speaker past it's limits causes driver distortion and power compression, which will result in mechanical failure and possibly heat failure, but that doesn't mean distortion had anything to do with it, as you never said the amp was beyond it's limits.

The only way to possibly make that out would be:

pushing amplifier past its limits = clipped wave = added TRMS power = more heat = possible thermal damage to speaker

There are a bunch of 'if's' packed into that though.

 

[chris229 10+ year member]

guys-------distortion doesn't kill speakers

amps DON'T draw current like blow driers or home appliances.

amps DON'T output power like blow driers or home appliances

speakers take the same amount of power clipped OR clean power to blow them

clipping isn't bad-----------------just don't do it in excess cause it sounds like crap

 

[jlaine 10+ year member]

Originally posted by chris229  

speakers take the same amount of power clipped OR clean power to blow them

 

Actually... If we want to get really technical..

Clipped waves reduce amount of cone excursion happening by truncating the peak of the waveform, or in essence "holding" the woofer at the excursion point, which reduces the amount of airflow going into the pole piece, which will in turn reduce the amount of thermal dissipation that the woofer has, which in turn will reduce the overall thermal handling that the unit has...

So under duress, the clipped wave is worse than the clean one, but both should still have to be above the rated RMS value of the woofer.

//content.invisioncic.com/y282845/emoticons/smile.gif.1ebc41e1811405b213edfc4622c41e27.gif

 

[MikeofTulsa Moderator]

here we go again.....

 

[Plater 10+ year member]

Hey buddy if read what i am saying you would fine that it is all factual.

you can't send 20amps to a speaker at 5 volts DC

The impedance of a speaker in usally higher in then the dc resistance but not by much

if a speaker is 4 ohms then a typically dc resistance is 3.6ohms if i put 5 volts across the terminals i will get current flow proportional to the resistance 5/3.6=1.36amps.

The coil satuartes and the speaker pushs in or out depeneding on the polarity

Anyways i am done arguing with you about this you seem to know everything ( jlaine).

 

[jlaine 10+ year member]

Originally posted by Plater Hey buddy if read what i am saying you would fine that it is all factual.

 

you can't send 20amps to a speaker at 5 volts DC

 

The impedance of a speaker in usally higher in then the dc resistance but not by much

 

if a speaker is 4 ohms then a typically dc resistance is 3.6ohms if i put 5 volts across the terminals i will get current flow proportional to the resistance 5/3.6=1.36amps.

 

The coil satuartes and the speaker pushs in or out depeneding on the polarity

 

Anyways i am done arguing with you about this you seem to know everything ( jlaine).

I cannot provide 20A to a speaker at 5vdc? Why not?

I was assuming a ridiculously small DCR, because that is what I need to get your current figures. And stay within the performance envelope I was given.

200watt amp, providing 20 amps of current (10 per channel), that is what I read. Did I read wrong? I may have, but I thought this was the discussion given was it not?

Where did I miss anything? The rest was nothing more than reiteration to prove a point, your 20 amps of current is incorrect. I never mentioned the speaker DCR, as to even get close to 20 amps we need to use a ridiculously small impedance with a high wattage amplifier.

Now, expanding on your own usage of ohms law... (we are using your 200 watt amp here)

DCR = 3.6 ohms.

Amperage = 10A

E = I*R right?

E = 10 * 3.6

E = 36V

See a problem with this? I do... It will take thirty-six volts!! to pass 10 amps of current through that coil!

Take that, and use P = V^2/R and see what the formula gives you for wattage....

P = 36^2/3.6 = 360 watts! And this is only for ONE SIDE add them both up, to get 720 watts...

Now, via ohms law, the current in the coil of the speaker is the amplifier output voltage, divided by the resistance of said coil at whatever frequency we are dealing with. Note how we calculate current is to base it upon the voltage.

60V/4 ohm coil. (we'll skip the impedance at the exact frequency for now, just because)

15A

P = V^2/R

P = 60^2/4

P = 900W

Just for 15 amps! Just for a single channel! I'd need a 1800Wrms amp to even touch 30 amps of current through a 4 ohm load. Drop the load, and assume current goes up, but what happens when the output fet's cannot keep up with the current demand?

A few things we haven't added to the fray, that need to be.

Inductance

Impedance @ frequency

Your 200W amp? Could do 20A of current...

If it could produce 80V into a 4 ohm load... Makes it much larger than 200W

If it could produce 40V into a 2 ohm load... Makes it much larger than 200W

If it could produce 20V into a 1 ohm load... Makes it much larger than 200W

lets assume it can do a 1 ohm load (easiest to work with here)

P = V^2/R

P = 20^2/1

P = 400

A pure 400W rms, 1 ohm stable amplifier. Double what you suggested, at a quarter of the resistance we were initially dealing with.

You still didn't answer my question... Where do you live? I've never heard of a 110A service..

Settle down plater, I'm not after you, just don't agree completely with what was stated.

 

[Plater 10+ year member]

The original speakers used for the 200W per channel amp so really 400W was 2 ohm loads not 4 ohm

So

200W=I^2*R

200/2=I^2

sqrt(100)=I which is 10amps

A 2ohm stable 2channel 400W amp 200W per channel is not uncommon

As the impedance of the speakers is dropped the less voltage gain is needed and more current gain is needed.

Watch this

200=V^2/R which is

200*2=V^2

400=V^2

SQRT(400)=V

V=20

Or

P=20*10(I*E)

P=200W

Or

P=I^2*R

P=10^2*2

P=200W

Or

P=V^2/2

P=20^2/2

P=200W

P=I*V is the main eq'n subbing in ohms law solved for I or V in terms of R gives the other equ.

All three variations of the power eq'n will work.

The fact is as the load impedance goes down the current draw goes up and the voltage needed across the load lowers to have the same power at the load.

This is why it is hard to make a power supply the will work well under different loading conditions. IE. An amp that can run loads from 1ohm to 16ohms. Because the current needs and voltage needs to get the same amount of power to the speaker differ greatly.

Anyways I would say that we are both right but we are getting caught up in the wording of things. Oh and you where right about current services to houses. But mine is 110amps used to be 50A old farmhouse the 100A and 200A service are more common but a difference of 10A come on. LOL.

One more thing most car audio amplifiers use BJT's as output devices and MOSFET's for the switching power supply part of the amp. I do know that higher power amps use MOSFET's for the outputs (because of easier paralleling positive temperature co etc.) but in general most amps are BJT's because they are cheaper and easier to get complementary pairs, when the are used in lower power apps.

When you built the amp you where talking about how did it work?

Was it a home or audio amp?

 

[jlaine 10+ year member]

Originally posted by Plater When you built the amp you where talking about how did it work?

Was it a home or audio amp?

Home or audio amp? Or did you mean home or car?

It's home audio, nothing off the wall, fairly standard TO-3 case fet's, unregulated power supply, 75wpc design. 35VDC power rails, benchmarked at 90wpc @ .01% thd @ 60hz. 24000uF supply bank, very low ripple design, 20~22K response, 4 ohm stereo, non-bridgeable.

Man, that's been collecting dust for a while, some of my specs may be off from my memory being sketchy... I built it... 7 years ago... I could bench it again in a few days if you like, it's around my house somewhere.

Back on track...

P = V^2/R is used to bench test amplifiers, always is.

Few reasons...

1. No need to be in the circuit path. (parallel connection, not series)

2. More accurate, clampmeters are too sensitive at such marginal loads.

Other thing disputed:

Gains are always voltage controls, the current is based on the load. The gains will try and set the voltage at the level they were designed to, reguardless of the load presented. That is why they are called such, and why they are seen as marked out in dBV.

The amplifier tries to maintain VOLTAGE levels at varying impedances, not current levels. That is why it is done the way it is. (until we get into dual power rails, like soundstream rubicon designs and JL's RISC setup)

If you drop the load, and leave the gain identical, the amplifier is going to TRY and present the same amount of voltage, with the higher current. Sometimes it can pull it off, other times something goes goofy.... //content.invisioncic.com/y282845/emoticons/smile.gif.1ebc41e1811405b213edfc4622c41e27.gif

 

[MikeofTulsa Moderator]

see what you started corsica...heh...interesting tho

 

[chris229 10+ year member]

Originally posted by jlaine  

 

Actually... If we want to get really technical..

 

Clipped waves reduce amount of cone excursion happening by truncating the peak of the waveform, or in essence "holding" the woofer at the excursion point, which reduces the amount of airflow going into the pole piece, which will in turn reduce the amount of thermal dissipation that the woofer has, which in turn will reduce the overall thermal handling that the unit has...

 

So under duress, the clipped wave is worse than the clean one, but both should still have to be above the rated RMS value of the woofer.

 

//content.invisioncic.com/y282845/emoticons/smile.gif.1ebc41e1811405b213edfc4622c41e27.gif

well not really------------------the signal is still whatever HZ and the cone is still moving the alot of air.

 

[Plater 10+ year member]

Sorry about the typo

You are right about using voltage and impedance when measuring power to a speaker because it is easier. But if you have both those values current drawnin can be calculated.

That sounds like a fairly nice little amp design.

Did you have any problems with the fet outputs. Biasing or stablity? Also what where the FET's you used, i am finding it pretty hard to find complemantary pair power MOSFET's. It seems most manf. want to just make there FET's for switching and mostly N-channel.

I don't know if you care about his or not but i was talking to a engineer at IRF today about a problem i was having with a 1Kw load box i am building at work. I am using 6 IRC740 fet with a hexsense current leads to use as feedback to regulated the current flowing in the devices. I also have them mounted to a .5 termal resistance sink and have 300cfm following over them. But anyways the gates kept being destroyed if i try to load suppies over 25 volts, it works lovely under 25V. He told me FET's are the wrong choice they are for switching and they won't share current well in the design i am using. Anyways!

You don't have to test your amp for me but thanks for the offer.

 

[jlaine 10+ year member]

Plater, I will have to get back to you on the FET's used, I have the part numbers here somewhere, I'll post 'em up ASAP... Stability- wise, it is fairly solid, not too finicky compared to some of what I've seen.

Something that has caught my eye is the velleman amp kits, they have a nice looking amplifier for dirt cheap if one was so inclined. A few people I talked to mentioned oscillation problems with the design, which can be resolved by using low capacitance wire usually, but there is also a few workarounds for it too.

 

[Steven 10+ year member]

Most of the amps i work with have markings on the gain controls.

markings like 0.2V, 0.5V, 2V, 4V

Last I knew V=Voltage

using this I'm guessing that the gain controls the voltage. I also know that if your Headunit puts out a 2V signal you should set your gain on your amp to a little more than 2V.

if you hook a multi-meter up to an amps outputs you'll find puts out both varying voltage and amperage.

Watts=Amps X Volts

Now one of my amps running at full power puts out 400watts per channel. When I put a voltmeter accross the terminals I get a reading of about 23V, The ammeter gets a reading of about 17.3 Amps. 23V X 17.3A = Approx. 400watts.