DiyMobileAudio.com Car Stereo Forum banner

1 - 20 of 20 Posts

·
Registered
Joined
·
1,978 Posts
Discussion Starter #1
got this really informative email from TIPS after asking if I could mod a us amps 2000x to run at 2 ohm mono

DUDE IS WAAAAY COOL for writing all of this--

thought some of you guys would find this information useful

A 2000X WILL NOT run below 4 ohms bridged, sorry!! If you wanted it to do this you should have bought a 2000 not a 2000x. There are limits to how much power a transistor or set of transistors can make, and the 2000 is at its limit into a 2 ohm speaker (1 Ohm bridged load) and a 2000x is at its limit into a 4 ohm speaker (2 ohm bridged load).

A 2000x is a high voltage designed amp, a 2000 is a high current designed
amp. They both make the same power just in different impedance loads.

Power is derived from voltage and current, thru physics and Ohm's Law.
Voltage and current also have INVERSE relationships, meaning that as voltage
increases current decreases to make the same amount of power. So if voltage
is low, current has to be high to make power (high current amp) when voltage
is high, current goes low (high voltage amp).

There is a line that you cannot cross or cross for long when making an amp
do a mixture of both high voltage and high current. If you cross that line
or come too close to it, the amp will destroy itself sooner or later, if not
immediately, and with the people that have cut the protection diodes out of
2000x's they have seen a VERY nasty failure! The circuit board is usually
burnt through in very large places making the amp a boat anchor, or door
stop.

High voltage amps are the most logical way to build an amp, you can make
just as much or more power from high voltage as you can from high current
WITH MANY LESS BAD EFFECTS. High currents amps only came into existence as cheater amps for contest purposes, (small power figure when rated at 4 ohms, but when ran into low impedance much more power made).

The 2000x and the 2000 are examples of this, they both make the same power but at different impedance loads. There was really no reason to make the 2000 if the 2000x made the same power, unless you wanted to cheat and and say your amp was only a 500 watt stereo amp at 4 ohms but get 2000 watts briged into 2 ohms.

The PROBLEM with high current (any brand of high current) is

1) twice as much heat is generated from the amp when the impedance is cut in half from say 4 ohms to 2 ohms, or 2 ohms to 1 ohm and so on

2) distortion is increased if not doubled every time the impedance is
reduced

3) current draw from your battery system is doubled every time the impedance is reduced (you have to have bigger, better and more batteries, just to make the same power, and if you don't and starve the amp for power the power supply in the amp will blow up)

4) the life of the transistors are reduced (basically cut in half) every
time the impedance is reduced, since they have to put out twice the power
(until you reach their manufactured limits then they just blow up)

5) every time you reduce the impedance the amp THEORETICALLY (paper watts)
makes more power, but they don't, as the impedance decreases the efficiency
of the transistor is decreased, and more power is used up as thermal energy
(heat) not acoustical energy (power to speakers) The power is not
neccessarily usable power to the speaker, yes the transistor is working
twice as hard and producing twice the power, but a larger precentage of this
power is be produced as heat which does nothing to move the speaker just
keep you warm in the winter, or hotter in the summer, ha ha.

6) headroom (reserve power) is cut in half every time the impedance is
reduced

7) damping factor (speaker control) is cut in half every time the impedance
is halved

High voltage amps avoid all of these problems and give you better sounding
power, more effieciently with a much longer amp life, and still can give you
high power outputs. It all boils down to a simplified mathmatical equation
from Ohm;s Law Volts x AMPs = Watts (this is very simpilfied but give you
the general formula)Knowing math you can see that you could raise the
voltage or the amperage figure to increase your power (watts) When building
amps you have to choice the proper mixture of these in order to make a amp
last, because the transistors that an engineer chooses can only make a given
amount of power and if you increse the votlage figure or the current figure
too close too or beyond the capability of the transistor it WILL DIE, maybe
not instantly but it will die, it depends on how bad you violate the rules.
ALL TRANSISTORS have their limits and specifications, and all will fail if
taken past their limits.

A 2000 uses an output transistor that is suited for high current loads but
not extreme high voltage, the 2000x uses a transistor that is suited fro
high voltage applications but not high current, other than that the circuit
boards are the same between the two models of amp, other than the power
supply transformers are wrapped a different number of time (turns of copper
wire) to make a certain voltage. So as you can see you cannot marry these
two designs together and make them work. So people try to run 2000x like
2000's and they may last a short period, as short as one contest sometimes)

This is why you see people crazy about amps like Linear Power, they were
HIGH VOLTAGE amps, they were extremely over built, and they last almost
forever. Again thats why you see some on these sites for sale that are 10,
15 and 20 years old still working and still being sold for major amounts of
money.

The only reason Linear Power amps can be modified (and last) is that they
are so overbuilt, and that we DO NOT make high current amps from them, we
improve thier high voltage characteristics. WE WILL NOT do a mod to any amp
that effects its life or sound quality, all of our mods are improvements in
power, sound quality, with no decrease in longivity or reliablity and in
most cases we are able to increase that too.

This is why you see such posts from individuals that commend our work, or
dedication, and that commend Linear Power products. We will not compromise
and make a product unstable or unreliable.

Thanks for asking anyway, sorry for the long response, but there is an awful
lost of explaination to be made to have an understanding if why I say no.
 

·
Registered
Joined
·
2,241 Posts
LP kicks your ass... and your ass (pointing elsewhere), and your ass, and they like me ... IM OUT! (throws mic)
 

·
Registered
Joined
·
1,978 Posts
Discussion Starter #3 (Edited)
LOL --makes me want one after seeing these pictures of a 2.2HV

found these pics and these specs

they kind of look like my old lanzar optidrve 2500 from the inside, powersupply looks the same, and the wires all over the place lol, to3's are cool too

Power Ratings for Linear Power 2.2HV

Specs are from the manual, their website has different specs listed which may be more indicative of the latest model.

12.5 x 2 watts @ 4 or 2 ohms (rated power)
135 x 2 watts @ 4 or 2 ohms @ 12.5v RMS
185 x 2 watts @ 4 or 2 ohms @ 14.4v RMS
250 x 2 watts @ 4 or 2 ohms @ 14.4v RMS with music power
425 x 1 watts @ 4 ohms or 2 ohms @ 12.5v
600 x 1 watts @ 4 ohms or 2 ohms @ 14.4v
750 x 1 watts @ 4 or 2 ohms @ 14.4v RMS with music power

40-amp external AGU fuse (included in the package)
 

·
Registered
Joined
·
17,149 Posts
newtitan said:
got this really informative email from TIPS after asking if I could mod a us amps 2000x to run at 2 ohm mono

DUDE IS WAAAAY COOL for writing all of this--

thought some of you guys would find this information useful
That is absolutely the most straight-forward, no-********, explanation as to what happens. I applaud them for stepping outside the "run the hell out of it" box and telling you this. Certainly a keeper, if not a wall decoration for my office :)

Chad
 

·
Registered
Joined
·
6,851 Posts
There's nothing wrong with "high current" amplifiers if designed correctly. Gotta give the guy credit for spending so much time to write to you (benefit of buying from a smaller company, huh?). But his list is completely bogus:

"1) twice as much heat is generated from the amp when the impedance is cut in half"

But doubling the voltage instead (to achieve the same power) will still generate that heat.

"2) distortion is increased if not doubled every time the impedance is
reduced"

This is a common misconception. Output stage distortion is only a part of the overall distortion in most amplifiers, and even then it's usually dominated by crossover distortion, or even gm-doubling (in the case of class AB). Poor bias tracking in most car audio amplifiers I've looked at is probably responsible for most of the OPS distortion anyway, but others may have a different viewpoint on this. Also, if the outputs are FETs or hybrids, a lot of the distortion that would arise from decreased impedances wouldn't be present because only BJTs show beta droop.

"3) current draw from your battery system is doubled every time the impedance is reduced (you have to have bigger, better and more batteries, just to make the same power, and if you don't and starve the amp for power the power supply in the amp will blow up)"

The same is true for higher voltage. Gotta get the power from somewhere.

"4) the life of the transistors are reduced (basically cut in half) every
time the impedance is reduced, since they have to put out twice the power
(until you reach their manufactured limits then they just blow up)"

Not necessarily. In fact, I'd argue the opposite. You're more likely to reach secondary breakdown with higher voltages. Again, since we're talking the same amount of power here, heat dissipation and thermal cycles won't be much different in the two cases.

"5) every time you reduce the impedance the amp THEORETICALLY (paper watts)
makes more power, but they don't, as the impedance decreases the efficiency
of the transistor is decreased, and more power is used up as thermal energy
(heat) not acoustical energy (power to speakers) The power is not
neccessarily usable power to the speaker, yes the transistor is working
twice as hard and producing twice the power, but a larger precentage of this
power is be produced as heat which does nothing to move the speaker just
keep you warm in the winter, or hotter in the summer, ha ha."

Is this true? Reduced efficiency of the transistor? With a BJT, this would manifest itself as beta droop. What about an FET? I wouldn't think so. What do you guys think? [Besides, you also get losses when you increase the supply voltage.]

"6) headroom (reserve power) is cut in half every time the impedance is
reduced"

Ditto for increasing voltage.

"7) damping factor (speaker control) is cut in half every time the impedance
is halved"

Damping factor is irrelevant in solid state designs utilizing global negative feedback, as shown by Dick Pierce and Doug Self, among others.
 

·
Registered
Joined
·
6,851 Posts
newtitan said:
LOL --makes me want one after seeing these pictures of a 2.2HV

found these pics and these specs

they kind of look like my old lanzar optidrve 2500 from the inside, powersupply looks the same, and the wires all over the place lol, to3's are cool too
Yeah, I was wondering about that! How many car amps have TO3 package transistors these days? I hate TO3Ps and it seems like I'm seeing more of those in car amps these days. At least TO220s don't break when you try to use the mounting hole (oops! :) ).
 

·
Registered
Joined
·
1,978 Posts
Discussion Starter #8
those wonderful Abyss (only in korea ;) ),


woops I mean TRU Technology (only in america :p ) amps are using them as far as I know off
 
W

·
Guest
Joined
·
0 Posts
I largely agree with Mark. In my opinion, the distinction between "high voltage" amps and "high current amps" is artificial. Here's why :

1. Almost all audio power amps are designed to mimic low-output impedance, voltage sources. They are designed for a certain impedance range. There's nothing fundamentally distinct about 16 ohms, 8 ohms, 4 ohms, or 2 ohms.

2. For any audio power amp, no matter if it's advertised as "high voltage" or "high current", when you drop the impedance all the things described do happen : output power increases, current demand increases, amp draws and dissipates more power. These are simple consequences of the "voltage source" nature of the amp ... and they all happen for "high voltage" amps as well.

3. An amp properly designed to deliver a rated power into a 2 ohm load, will get no hotter or break any sooner than an amp properly designed to deliver the same power into a 16 ohm load ... providing, of course, the amps are the same topology class. Remember that efficiency calculations for Class A, Class AB, etc. do not depend fundamentally on the load impedance (see point number 1).

No question, Linear Power made great amps. They were optimized for impedances on the "high side" of what's typical in car audio ... that's all.
 
W

·
Guest
Joined
·
0 Posts
oh by the way ... I don't understand the concept of "efficiency of the transistor". Perhaps he meant gain of the transistor?

Efficiency is typically defined simply as output power divided by input power. Certainly has meaning for a power amp, input power and output power are readily calculated ... the difference is of course dissipated as heat by the amp. And yes, that heat is dissipated by the output transistors.
 

·
Registered
Joined
·
17,149 Posts
In many measured instances an amplifier's efficiency will decrease along with applied impedance. The amplifier not the transistor.

Power is power be it voltage or current driven, higher voltages damage transistors as fast. You will most likely see voltage ratings on transistors (power) before wattage ratings.

Dropping the rail voltage makes it easier for a car amp to make power, it will still make the same max potential power regardless, lower voltages makes it do it easier.

I have a cashed Phoenix gold M275 here, PS is still good, it has a "high current mode" I'll drag it out soon and report back as to what is actually happening inside.

There is way too much hype put into bridging and low impedance loading. Buy the power you need and load it like a nice person and it will last a long time.

Chad
 
W

·
Guest
Joined
·
0 Posts
It's power that causes irreversible damage to semiconductors ... voltage or current alone will not do it.

If a transistor experiences a high-voltage breakdown mechanism, it will not damage the device unless the breakdown causes enough current flow that the device can no longer dissipate the heat. Junction breakdown mechanisms happen all the time ... many devices (like Zener diodes) make use of them. They are "reversible", in the sense that reducing the voltage will return the device to normal operation.

In this regard, it's power that's fundamental ... not voltage or current alone.

Power kills !! :)

Just for completeness, there a few potentially damaging mechanisms that depend on current alone ... electromigration comes to mind. But I doubt that a well-designed audio power amp has ever failed due to this effect ;)
 

·
Registered
Joined
·
553 Posts
I've got several modded Linear Power amps from Ray and they are easily the best amps I've ever owned. I'm glad he's still around. Too bad LP is gone.
 

·
Registered
Joined
·
6,851 Posts
werewolf said:
It's power that causes irreversible damage to semiconductors ... voltage or current alone will not do it.

If a transistor experiences a high-voltage breakdown mechanism, it will not damage the device unless the breakdown causes enough current flow that the device can no longer dissipate the heat. Junction breakdown mechanisms happen all the time ... many devices (like Zener diodes) make use of them. They are "reversible", in the sense that reducing the voltage will return the device to normal operation.

In this regard, it's power that's fundamental ... not voltage or current alone.

Power kills !! :)

Just for completeness, there a few potentially damaging mechanisms that depend on current alone ... electromigration comes to mind. But I doubt that a well-designed audio power amp has ever failed due to this effect ;)

For the most part, I agree with you. But to be nitpicky, which is my way :) , the SOA curves of a "typical" bipolar transistor will usually be asymmetrical with respect to the current and voltage axes. The secondary breakdown curve is one classic example of this. As a result, the linear relation (which isn't perfectly linear to begin with) tends to crap out at high voltages.

Anyway, what I'd like to see is a (good) analysis of the benefits/drawbacks of using either bipolar or FET output transistors for high voltage or high current applications. My inclination is to say that there would be a difference, however minor, and that high current applications would benefit by using FETs (even if it's with bipolar drivers). Aside from being more robust (arguable point, I suppose), they're resistant to beta droop which is especially prominent with high current draw. Does anyone know if Vce tends to be higher than Vds in a "typical" set of power transistors? I would think that would factor in as well.
 

·
Registered
Joined
·
6,851 Posts
chad said:
In many measured instances an amplifier's efficiency will decrease along with applied impedance. The amplifier not the transistor.
But does it decrease in comparison with the higher voltage counterpart? I would assume that the efficiency would decrease equally in either case (with the possible exception of whatever's lost across emitter resistors or zobel networks, etc, which I would assume would be minimal).
 
W

·
Guest
Joined
·
0 Posts
Whenever power is transferred or applied at a higher current level, rather than a higher voltage level, you are potentially subject to higher (I^2)*R losses, which can potentially degrade measured efficiency into lower impedance loads.

I've used the word "potentially", because proper design of the whole high-power path ... including connectors and speaker wire, of course ... can address this issue very effectively.

You know, it often helps to think of "scaling" arguments to demonstrate a point. Imagine that an amplifier is perfectly happy driving an 8 ohm load. Suppose I've got four of these amp/load combos, each applying the exact same signal to their respective 8 ohm loads (this is a "thought experiment" for now, where we can imagine the gain settings for each amp are identical, and they all receive the same input signal).

Now I connect all four loads in parallel, still allowing each amp to drive it's load (Yes ... the amp outputs are tied together now, something you would not necessarily want to do in a real experiment :( ) The real load is now 2 ohms. But has efficency, damping, anything else changed?
 

·
Registered
Joined
·
17,149 Posts
MarkZ said:
But does it decrease in comparison with the higher voltage counterpart? I would assume that the efficiency would decrease equally in either case (with the possible exception of whatever's lost across emitter resistors or zobel networks, etc, which I would assume would be minimal).

Dunno, I just fix em :)

I do know that the efficiency of an amp will pitfall as impedance drops. I've never had the urge to kick up the rails, but I do know in RF applications efficiency does rise with voltage, albeit this is a fixed impedance output!

Coming from the class "G" thread, runing them at a lower voltage helps their efficiency and raising the voltage for the peaks decerases it but overall efficinecy is raised :)

I still feel that regardless, running amps at insanely low impedances is asking for trouble, but, puts money in my pocket :)

Chad
 

·
Registered
Joined
·
17,149 Posts
werewolf said:
It's power that causes irreversible damage to semiconductors ... voltage or current alone will not do it.

If a transistor experiences a high-voltage breakdown mechanism, it will not damage the device unless the breakdown causes enough current flow that the device can no longer dissipate the heat. Junction breakdown mechanisms happen all the time ... many devices (like Zener diodes) make use of them. They are "reversible", in the sense that reducing the voltage will return the device to normal operation.

In this regard, it's power that's fundamental ... not voltage or current alone.

Power kills !! :)

Just for completeness, there a few potentially damaging mechanisms that depend on current alone ... electromigration comes to mind. But I doubt that a well-designed audio power amp has ever failed due to this effect ;)

Transistors have power dissapation and voltage maximum ratings. What you said is true BUT, if the transistor goes into voltage breakdown in a class A application you simply have DC, if it does it in AB or B the flyback diodes will cause it to make incredible current, fast. Thus causing thermal failure. So... yes it's thermal failure but the overvoltage caused it. Overshoot gets wicked too :) Don't look at the transistor but also the support components as a system, things DO get ugly quick.

Chad
 

·
Registered
Joined
·
6,851 Posts
chad said:
Dunno, I just fix em :)

I do know that the efficiency of an amp will pitfall as impedance drops. I've never had the urge to kick up the rails, but I do know in RF applications efficiency does rise with voltage, albeit this is a fixed impedance output!

Coming from the class "G" thread, runing them at a lower voltage helps their efficiency and raising the voltage for the peaks decerases it but overall efficinecy is raised :)

I still feel that regardless, running amps at insanely low impedances is asking for trouble, but, puts money in my pocket :)

Chad
So the class G example demonstrates that the efficiency is decreased with increased rail voltage, since the voltage across the transistor is going to be Rail minus voltage from the driver minus the Vbe. So that means the power dissipation in the transistors would increase by sqrt(2)*X, except when the signal exceeds X. How much this would change the overall efficiency, I don't know, and how that compares to the half-impedance load case remains a mystery. I may have to do some digging for clues. I'll let you know if I turn up anything.
 

·
Registered
Joined
·
17,149 Posts
MarkZ said:
So the class G example demonstrates that the efficiency is decreased with increased rail voltage, since the voltage across the transistor is going to be Rail minus voltage from the driver minus the Vbe. So that means the power dissipation in the transistors would increase by sqrt(2)*X, except when the signal exceeds X. How much this would change the overall efficiency, I don't know, and how that compares to the half-impedance load case remains a mystery. I may have to do some digging for clues. I'll let you know if I turn up anything.

At this point efficiency depends on signal input duty cycle, at sine testing it would be easier to calculate but the dramatic efficiency increase is due to the application's crest factor.

At one point I considered a JL Audio 300/4 I'll try to dig up the e-mails back and forth as to how EXACTLY it senses high current /vs/ High voltage mode and how fast, it' an interesting read. Ended up NOT being the amp for me.

Chad
 
1 - 20 of 20 Posts
Top