Was: Dummy load to keep tube amp happy
By Rachel007 - usenet poster
One loudspeaker connection to my much cherished ST70 became, unbeknown
to me, disconnected. Result, one fried output transformer!
Would a dummy load resistor, as mentioned by Robert Casey, give protection?
What would be the procedure if the amplifier only had one set of output
terminals, such as VTL, etc?
Robin
Solution #1
posted on Aug 02, 2007
Ross - usenet poster
Rank:
Apprentice
Rating: 0%, 0 votes
ApprenticeRating: 0%, 0 votes
In <460lpc$@canyon.sr.hp.com
Guitar amp manufacturers frequently *do* address this, in a couple
different ways. I guess hifi manufacturers assume your speakers
stay connected.
Fender amps (and others) use shorting output jacks. Not sure what
the load line looks like with a dead short, but the theory is probably
'tubes are cheaper than transformers'. Of course, this will not help
if a speaker cable is plugged in with no speaker attached.
Someone (Dave Cigna, I believe) said his Traynor amp uses a 100 ohm,
10 watt resistor internally. The wierd thing: apparently they use
a shorting output jack as well, which would indicate that they didn't
think the 100 ohm load was always sufficient protection. Unlike
Marshall and some others, Traynors were reputed to use good quality iron.
Some amps (newer Fenders incl'd I believe) use hivoltage rects from
output tube plates to ground. By the way, I never trust diode markings,
so when I do this I always measure the diodes first. Obviously putting
them in the wrong way or putting in shorted diodes will sort of defeat
the intended purpose (!)
Dexter McNeil made the interesting suggestion of using TranZorbs (like
huge zeners) in a similar config.
[good circuit analysis snipped...read Kurt's post]
Keep in mind that most tube amps are used by guitarists. Sane?
You may be stretching the definition <g This may expain the discrepancy between your calc'd 1W sane resistor
value and the 10W resistor in the Traynor. I'll stick with the higher
wattage resistors in my guitar amps for now. Besides, I always like an
excuse to bolt one of those shiny big metal guys to the chassis. <g
Yep, hardly even feel it. Probably a worthwhile investment for the
cautious, or those who disconnect and reconnect speakers frequently.
Another interesting observation (by Henry I believe?) is that old
table radios and other single-speaker systems used caps/resistors
across the output transformers...cause a single speaker (guitar amps
again) will exhibit high imp. peaks which could also look close to
an open circuit at some frequencies. Doubt that this is a problem
in multi-speaker systems.
I'm still curious about the post that mentioned a 1k feedback resistor
getting fried because the speaker was not plugged in. I'm guessing that
the resistor may have been fried beforehand (?)
Regards,
Mark Garvin
Guitar amp manufacturers frequently *do* address this, in a couple
different ways. I guess hifi manufacturers assume your speakers
stay connected.
Fender amps (and others) use shorting output jacks. Not sure what
the load line looks like with a dead short, but the theory is probably
'tubes are cheaper than transformers'. Of course, this will not help
if a speaker cable is plugged in with no speaker attached.
Someone (Dave Cigna, I believe) said his Traynor amp uses a 100 ohm,
10 watt resistor internally. The wierd thing: apparently they use
a shorting output jack as well, which would indicate that they didn't
think the 100 ohm load was always sufficient protection. Unlike
Marshall and some others, Traynors were reputed to use good quality iron.
Some amps (newer Fenders incl'd I believe) use hivoltage rects from
output tube plates to ground. By the way, I never trust diode markings,
so when I do this I always measure the diodes first. Obviously putting
them in the wrong way or putting in shorted diodes will sort of defeat
the intended purpose (!)
Dexter McNeil made the interesting suggestion of using TranZorbs (like
huge zeners) in a similar config.
[good circuit analysis snipped...read Kurt's post]
Keep in mind that most tube amps are used by guitarists. Sane?
You may be stretching the definition <g This may expain the discrepancy between your calc'd 1W sane resistor
value and the 10W resistor in the Traynor. I'll stick with the higher
wattage resistors in my guitar amps for now. Besides, I always like an
excuse to bolt one of those shiny big metal guys to the chassis. <g
Yep, hardly even feel it. Probably a worthwhile investment for the
cautious, or those who disconnect and reconnect speakers frequently.
Another interesting observation (by Henry I believe?) is that old
table radios and other single-speaker systems used caps/resistors
across the output transformers...cause a single speaker (guitar amps
again) will exhibit high imp. peaks which could also look close to
an open circuit at some frequencies. Doubt that this is a problem
in multi-speaker systems.
I'm still curious about the post that mentioned a 1k feedback resistor
getting fried because the speaker was not plugged in. I'm guessing that
the resistor may have been fried beforehand (?)
Regards,
Mark Garvin
Was this solution helpful? Show your Appreciation by rating it:
Solution #2
posted on Aug 02, 2007
Melissa - usenet poster
Rank: Apprentice
Rating: 0%, 0 votes
ApprenticeRating: 0%, 0 votes
In <4617bo$@hardcopy.ny.jpmorgan.com
If you are going to be running signal thru the tube amp, then don't
chance it. The 100 ohm (or whatever) should be regarded as a short-
term protective device. You may want to switch in another dummy load
at the same time you disconnect the speakers. Try not to short the
outputs of both amps together in the process, tho.
Or better yet, why not just run a tap from your video system into
the tube amp?
SS amps (w. no output transformer) don't have the same constraints.
But consider the other suggestion above.
Besides, we don't like none o' them solid state amps around these-here
parts <g
Regards,
Mark Garvin
If you are going to be running signal thru the tube amp, then don't
chance it. The 100 ohm (or whatever) should be regarded as a short-
term protective device. You may want to switch in another dummy load
at the same time you disconnect the speakers. Try not to short the
outputs of both amps together in the process, tho.
Or better yet, why not just run a tap from your video system into
the tube amp?
SS amps (w. no output transformer) don't have the same constraints.
But consider the other suggestion above.
Besides, we don't like none o' them solid state amps around these-here
parts <g
Regards,
Mark Garvin
Was this solution helpful? Show your Appreciation by rating it:
Solution #3
posted on Aug 02, 2007
Gary10 - usenet poster
Rank: Apprentice
Rating: 0%, 0 votes
ApprenticeRating: 0%, 0 votes
This discussion plays into a mini-project I've been thinking about: I need a
switch that lets me switch the main speakers between the tube-amp
(40wpc@8ohms) and the video SS system (100wpc@8ohms).
According to this discussion, perhaps all I need is to put in a
parallel resistor (say 56-68ohms) across the output of the amp? Then
use a 'break-before-make' circuit which is easy to build.
In that arrangement, however, the amp would then see an equivalent load
of 7.1 ohms (assumes 8 ohm speaker impedance {which further assumes
constant impedance which isn't really going to happen in the real
world}).
So my questions are:
With a 56-68ohm resistor in parallel with the speaker, are there
noticable effects on the characteristics of the speaker (frequency
response, etc.)?
Is a 56-68ohm resistor @ 20Watts the right size for the tube amp?
Use the same for the SS amp or use something else?
Thx,
Lewis
switch that lets me switch the main speakers between the tube-amp
(40wpc@8ohms) and the video SS system (100wpc@8ohms).
According to this discussion, perhaps all I need is to put in a
parallel resistor (say 56-68ohms) across the output of the amp? Then
use a 'break-before-make' circuit which is easy to build.
In that arrangement, however, the amp would then see an equivalent load
of 7.1 ohms (assumes 8 ohm speaker impedance {which further assumes
constant impedance which isn't really going to happen in the real
world}).
So my questions are:
With a 56-68ohm resistor in parallel with the speaker, are there
noticable effects on the characteristics of the speaker (frequency
response, etc.)?
Is a 56-68ohm resistor @ 20Watts the right size for the tube amp?
Use the same for the SS amp or use something else?
Thx,
Lewis
Was this solution helpful? Show your Appreciation by rating it:
Solution #4
posted on Aug 02, 2007
Reynolds - usenet poster
Rank: Apprentice
Rating: 0%, 0 votes
ApprenticeRating: 0%, 0 votes
: :
Ouch! There goes my theory you don't really need a protective resistor here.
So why don't these designers put the protection in there for you?
: : :
Yes, solder a resistor across the speaker output inside the chassis.
What value is right? Let's calculate it.
The effective primary load should equal the resistor times the ratio of the
primary impedance to the secondary impedance of the output transformer.
With a B+ of 450V and a power output about 40W, this transformer must
have a primary impedance of about 2500 ohms with an eight ohm secondary
load. That's an impedance ratio of 2500:8 or 313:1. The current needed
to produce 40W into 2500 ohms is 0.126 Arms.
Also, the transformer has a primary inductance with no load, and to pass
20 Hz with 2500 ohms reactance (-3dB) it must be 19.9 Henries.
If you put in a 100 ohm resistor internal on the secondary and no other
load connected, you will get an effective primary impedance of 313*100,
or 31K ohms. The circuit looks something like this, neglecting plate
resistance, which is high for these pentodes, but adding a nominal amount
of plate and primary and stray capacitance (just guessing at 1000 pF):
++---++
| | | |
I, max=0.126Arms 20H 1000pF 31Kohms
| | | |
++---++
For safe 900Vpp operation, current I must be less than 0.029Arms, but we
have capability for 0.126Arms. It could still overvoltage the transformer
and the output tubes, but we need to find the Q of the system to know how much.
Q = R*SQRT(C/L) = 0.22, which is small and not a problem. So we will be
clipping before we send transients into overvoltage.
If the load is 1K, the primary load is 310K, and Q = 2.2 which means we
will have transient overshoot and resonant peaking. This could overvoltage
the primary circuit and cause damage. 1K in this case is too high.
So, roughly, we could be protecting the amp with a 100 ohm resistor, but
not with an internal 1K feedback resistor. For a 40W amp, where you
would never deliver an *average* power more than 4W if you are sane,
the voltage across the load is 5.6Vrms. So the power into 100 ohms would
be 0.3W. You only need a 100 ohm 1W resistor soldered between COM and the
8 ohm speaker tap. You will lose 7% of your 40W power, or 0.6 dB, not
much.
Kurt
Ouch! There goes my theory you don't really need a protective resistor here.
So why don't these designers put the protection in there for you?
: : :
Yes, solder a resistor across the speaker output inside the chassis.
What value is right? Let's calculate it.
The effective primary load should equal the resistor times the ratio of the
primary impedance to the secondary impedance of the output transformer.
With a B+ of 450V and a power output about 40W, this transformer must
have a primary impedance of about 2500 ohms with an eight ohm secondary
load. That's an impedance ratio of 2500:8 or 313:1. The current needed
to produce 40W into 2500 ohms is 0.126 Arms.
Also, the transformer has a primary inductance with no load, and to pass
20 Hz with 2500 ohms reactance (-3dB) it must be 19.9 Henries.
If you put in a 100 ohm resistor internal on the secondary and no other
load connected, you will get an effective primary impedance of 313*100,
or 31K ohms. The circuit looks something like this, neglecting plate
resistance, which is high for these pentodes, but adding a nominal amount
of plate and primary and stray capacitance (just guessing at 1000 pF):
++---++
| | | |
I, max=0.126Arms 20H 1000pF 31Kohms
| | | |
++---++
For safe 900Vpp operation, current I must be less than 0.029Arms, but we
have capability for 0.126Arms. It could still overvoltage the transformer
and the output tubes, but we need to find the Q of the system to know how much.
Q = R*SQRT(C/L) = 0.22, which is small and not a problem. So we will be
clipping before we send transients into overvoltage.
If the load is 1K, the primary load is 310K, and Q = 2.2 which means we
will have transient overshoot and resonant peaking. This could overvoltage
the primary circuit and cause damage. 1K in this case is too high.
So, roughly, we could be protecting the amp with a 100 ohm resistor, but
not with an internal 1K feedback resistor. For a 40W amp, where you
would never deliver an *average* power more than 4W if you are sane,
the voltage across the load is 5.6Vrms. So the power into 100 ohms would
be 0.3W. You only need a 100 ohm 1W resistor soldered between COM and the
8 ohm speaker tap. You will lose 7% of your 40W power, or 0.6 dB, not
much.
Kurt
Was this solution helpful? Show your Appreciation by rating it:
Solution #5
posted on Aug 02, 2007
Reynolds - usenet poster
Rank: Apprentice
Rating: 0%, 0 votes
ApprenticeRating: 0%, 0 votes
Hi Robin, I believe the usual thing to do is to mount a higher value
resistor (100 ohms or so?) inside the amp chassis. Make sure wattage
is high enough.
Mark G.
resistor (100 ohms or so?) inside the amp chassis. Make sure wattage
is high enough.
Mark G.
Was this solution helpful? Show your Appreciation by rating it:
Suggest a new solution for this problem
Post a New problem for Ohm 4" 40W SHIELDED WOOFER, 8 Car
Email this problem
Post a New problem for Ohm 4" 40W SHIELDED WOOFER, 8 Car
Email this problem
Can you Help with these Car Speakers and Subwoofers problems?
Car Speakers and Subwoofers
Hi, How to find the Unlocking
Hi, How to find the Unlocking
Hi,
How to find the Unlocking... Answer This...
Car Speakers and Subwoofers
2004 Isuzu Rodeo factory in
2004 Isuzu Rodeo factory in
2004 Isuzu Rodeo factory in... Answer This...
Repair Service
Find Car Speaker and. Repairman Near You:
FixYa does not evaluate or guarantee the accuracy of any information provided through its proposed solutions,
posts, or Expert Assistance Sessions. By entering this site you declare you read and agreed to its Terms.
You may NOT copy or distribute the content that appears on this site without written
permission from FixYa Inc.
About | News
| Contact
| Find Products
| Find Cars
| Terms
| Privacy Policy
| FAQ
Tags | Blog | Repair Directory | Extended Warranty | Partners | Jobs |
Tags | Blog | Repair Directory | Extended Warranty | Partners | Jobs |