Impedance mismatch controversy
David Chertock
Years ago I learned everything I know about speaker / amp impedance
mismatches from the manual of my Groove Tube Speaker Emulator. I was
doing a google search about something else and came across numerous
a.g.a. postings as well as the FAQ that directly contradict what I had
read and believed.
The Groove Tube Speaker Emulator manual says:
"Your amp will most likey have an optimum operating output impedance
somewhere between 4 and 16 ohms. Our Speaker Emulator provides a load of
around 16 ohms. If your amp has a selectable impedance switch, best
results are achieved at the 16 ohm setting. However if your amp does not
have a selectable impedance, don't worry, it will work just fine. Always
remember never to load your amp with an impedance LOWER than its
designed for. For example, if your amp is a Fender Twin is which
designed to produce optimum performance into a 4 ohm load, a load of 2
ohms could be damaging. However a load of 8 or 16 ohms will result in a
loss of optimum power output, but will not harm the output stage. We've
found that the best results for a speaker emulator load device is around
16 ohms, so that's where we've set our impedance on the Speaker Emulator."
What the @#$%?
David
John Cox
You're riding a bike down a hill in a fairly comfortable gear.
All of a sudden you drop it into a lower gear and your legs can't go fast
enough to keep up. You're legs blow up from trying to keep up and you die
from loss of blood through those major arteries in your legs.
Ok, you're back to pedaling downhill in a fairly comfortable gear.
All of a sudden you kick the bike up to a higher gear and it gets harder to
pedal. This time you simply stand up to pedal and you accelerate without
having your legs blow up. Sure, it's a little more work, but you don't die.
Your amp works much the same way. It needs a fair amount of resistance (or
impedance in this case) to work properly. A little extra impedance is ok.
But too little resistance will make your amp try to "pedal" too fast. Get
it?
"David Chertock" <ap...@lafn.org> wrote in message
news:3E501AE5...@lafn.org...
JTM50
In reality going up or down on impedience "notch" typically isn't a big
concern. But going up is more of a concern than going down with tube amps.
The reason for this is becase of inductive kick back from the speakers.
It's the opposite for Transistor outputs which don't have transformer
coupled outputs.
Lloyd
in article 3E501AE5...@lafn.org, David Chertock at ap...@lafn.org wrote
on 2/16/03 3:12 PM:
JTM50
Weather you opt for a mismatch up or down is centered around inductive
kickback with tube amps (which you want to avoid) and current demamds with
transistors.
With a low mismatch (low impedience) kickback is minimalized, while with a
high mismatch kickback is increased (tubes with transformer outputs).
Why do you think that running a tube amp with out a load (infinate
impediance for a load) is a well known no no?
As a matter of fact some tube amps have jacks that short automatically
(extremely low impedience) when there isn't a plug in them. This saves the
amp from the potentially disatrous situation of the amp not having a load
(extremely high impedience).
On the other hand a high impediance load to the Solid state output just
limits current (and therefore the power to the speakers) while a low
impedience load or a short (an extremely low impedience) draws too much
current which will fry solid state components...
Lloyd
in article fkW3a.9263$jR3.4...@news1.news.adelphia.net, John Cox at
jo...@xpertimage.com wrote on 2/16/03 4:49 PM:
Phil Allison
"JTM50" <lgi...@telus.net> wrote in message
news:BA759357.17682%lgi...@telus.net...
> That just goes to prove that GT is a joke.
> In reality going up or down on impedience "notch" typically isn't a big
> concern. But going up is more of a concern than going down with tube amps.
> The reason for this is becase of inductive kick back from the speakers.
** A dummy load is not a loudspeaker - if it is equal to a resistor at
high frequencies then there is *no* inductive kickback. The amount of
kickback energy depends on the previous current level so high impedance
inductive loads have less.
............ Phil
Rich Koerner
Hmmmmmmmmmmmmmmmmmmm......................!!!!!!!!!!!!!!!!!!!!!!!!!
<thinking>
David Chertock wrote:
>
> Hi all,
> Years ago I learned everything I know about speaker / amp impedance
> mismatches from the manual of my Groove Tube Speaker Emulator. I was
> doing a google search about something else and came across numerous
> a.g.a. postings as well as the FAQ that directly contradict what I had
> read and believed.
>
> The Groove Tube Speaker Emulator manual says:
>
> "Your amp will most likey have an optimum operating output impedance
> somewhere between 4 and 16 ohms. Our Speaker Emulator provides a load of
> around 16 ohms. If your amp has a selectable impedance switch, best
> results are achieved at the 16 ohm setting.
Below are the first of two red flags, not to be missed.
> However if your amp does not
The use of the word "However", usually indicates a *wild card* has just
been played, to the previous statement. When the word *if* follows the
word "However", what follows is a total departure from the previous
statement.
Hmmmmmmmmmmmm....................!!!!!!!!!!!!!!!!!!!!!!!!
<thinking>
Notice now, the attempt of reconnection to the previous statement on
impedance relevance.
> However if your amp does not
> have a selectable impedance,
With this additional connection to a non selectable type of amplifier,
which is a common type of amplifier, you are now expecting some
important details for such to follow.
Well, guess what. They don't!!!!!!!!!!!!!!!!!!!!!!!!!!
Instead, they give the following.
> However if your amp does not
> have a selectable impedance, don't worry, it will work just fine.
There are those ever famous words, "Don't Worry, It Will Work Just
Fine", that have kicked so many people in the ass. Which are generally
followed by the words, "Oh Shit"!!!!!!!!!!!!!!!!!!!
When this happens, and someone hears the former words, you will usually
hear the words, "The time to WORRY, is when some Idiot says, Not
To"!!!!!!!!!!!!!!!!!!!!!!
Especially when, YOU don't fully understand the details of what is
involved.
GT has no understanding of what happens when back Electro Motive Forces
are created by mismatches in impedance in the vacuum tube amplifier at
different power levels of operation.
> Always
> remember never to load your amp with an impedance LOWER than its
> designed for. For example, if your amp is a Fender Twin is which
> designed to produce optimum performance into a 4 ohm load, a load of 2
> ohms could be damaging. However a load of 8 or 16 ohms will result in a
> loss of optimum power output, but will not harm the output stage.
Here, is seems they are only seeing the effects on the amp of the
current side of the problem of lower load mismatching. They are not
seeing the detailed effects on the amplifier from the voltage side of
the mismatching problem using higher loads than the amplifier's rated
load.
> We've
> found that the best results for a speaker emulator load device is around
> 16 ohms, so that's where we've set our impedance on the Speaker Emulator."
This statement creates the false notion, that it is OK to mismatch
vacuum tube amplifier loads across the board. Which is total
BS!!!!!!!!!!!!!!!!!!
The maximum SAFE transfer of energy, happens ONLY when the loading of
the coupling devices, are properly MATCHED to each other.
There is no exception, unless BOTH coupling devices are designed for
such.
Which, is clearly NOT, the case in point here.
Regards,
Rich Koerner,
Time Electronics.
http://www.timeelect.com
Specialists in Live Sound FOH Engineering,
Music & Studio Production,
Vintage Instruments, and Tube Amplifiers
elliott
> "JTM50" <lgi...@telus.net> wrote
> > In reality going up or down on impedience "notch" typically isn't a big
> > concern. But going up is more of a concern than going down with tube amps.
> > The reason for this is becase of inductive kick back from the speakers.
> ** A dummy load is not a loudspeaker - if it is equal to a resistor at
> high frequencies then there is *no* inductive kickback. The amount of
> kickback energy depends on the previous current level so high impedance
> inductive loads have less.
>
> ............ Phil
That's not been my experience. I've seen plenty of inductive spikes
into a resistive load with an upwards mismatch. Try this: set a
Marshall 100W to 4 ohms, connect it to a 16 ohm resistive load, dime it
out, play some chords into it, and scope the output. You will see
spikes.
Cheers,
elliott
Phil Allison
"elliott" <ell...@diespambotsdie.intouch.bc.ca> wrote in message
news:3E5084...@diespambotsdie.intouch.bc.ca...
** There are often MAJOR inductive spikes with a nominally **correctly
loaded** amp when run very hard into real speakers or simulated reactive
loads - I have measured up to 5 kV peaks on the tube plates with a 100 W
Marshall with 8 ohms plus 5 mH.
With resistive dummy loads such spikes are gone or relativly quite
small - they are now due only to the much smaller OT leakage reactance.
Now - what is your point ?
............. Phil
Phil Perkins
John Cox
amps. The analogy came out of experience with car stereos. Sorry for any
confusion. Thanks.
"JTM50" <lgi...@telus.net> wrote in message
news:BA7598B6.17683%lgi...@telus.net...
Miles O'Neal
> I was unaware. I'm still a little new to the actual inner workings of tube
> amps. The analogy came out of experience with car stereos. Sorry for any
> confusion. Thanks.
And for solid state equipment (such as car stereos)
it's a good analogy. But tube and SS gear are
backwards in this respect.
-Miles
elliott
Ok, try this one: Dime out your Marshall, play it into an open circuit,
and scope the output. Load inductance - zero. Load current - zero.
Should be fine by your reckoning, right?
When you get back to me with the results, you can place an order with me
for the new OP tranny, the set of EL34s, the 4 8pin sockets, and the 4
screen grid resistors that you'll need.
My point - you're full of it.
Cheers,
elliott
Phil Allison
** How smartarse. That is another situation.
............... Phil
elliott
> > > > > ** A dummy load is not a loudspeaker - if it is equal
> > > > > to a resistor at high frequencies then there is
> > > > > *no* inductive kickback.
Any competent tech on this newsgroup with a decent set of
resitive dummy loads can easily demonstrate that an upwardly
mismatched resistive load will result in inductive spikes.
Like I said, you're full of it.
Ken Gilbert
well, the spikes will be there, just as they are with a normal load,
but they will be larger in magnitude due to the lower Q of the
circuit.
ken
Phil Allison
** I have already answered this nonsense :
** There are often MAJOR inductive spikes with a nominally **correctly
loaded** amp when run very hard into real speakers or simulated reactive
loads - I have measured up to 5 kV peaks on the tube plates with a 100 W
Marshall with 8 ohms plus 5 mH.
With resistive dummy loads such spikes are gone or relativly quite
small - they are now due only to the much smaller OT leakage reactance.
Now - what is your point ?
>
> Like I said, you're full of it.
** You have no new point Elliot.
................. Phil