Help build a load box?

JakeAC5253 · Post by **JakeAC5253** » 12 Jul 2011, 20:18

A friend and myself have need or desire for a load box (no need for attenuation, simple passive dissipative load only) for our tube amps for the purpose of silent recording from the amp's FX send or line out. I have a resistor chosen which I believe is adequate, but I want to add heat sinks and fans as well. I am thinking that between the resistor, one heatsink, and a throughput fan system (one input fan and one output fan) it will be enough for sustained recording sessions of hours or more.

Here is the resistor I have chosen.
http://www.mouser.com/ProductDetail/Ohm ... 2bufagc%3d

And I have no clue about how to choose the right heatsink or fans, so I'm open to recommendations on those.

Also, I sort of want to figure out a way to thermally uncouple the resistor/heatsink arrangement from the encasement so that the case does not become a heatsink in and of itself. It could become quite hot to the touch during extended usage. Is this what I need to accomplish this?
http://www.mouser.com/ProductDetail/Cry ... 182HQsE%3d

phatt · Post by **phatt** » 12 Jul 2011, 22:15

Hi Jake,
If you need a fan you are doing it wrong.

Also, Tell us the wattage of the rig in question? That would help.

I very much doubt you would need a 300 Watt resistor?

Phil.

deltafred · Post by **deltafred** » 12 Jul 2011, 22:42

I worked with a guy who used to work at Marshall, he said when they were testing amps they had dummy loads that were just a sheet of expanded metal (steel) with a couple of wires bolted on with a jack plug and a scope lead breakout on the other end. I never asked him what the dimensions were.

rocklander · Post by **rocklander** » 12 Jul 2011, 22:47

years ago, a place I worked at we used a toaster element (cut the conductor to 8ohm I guess.. dunno.. never ran the impedance meter across it myself as it was there when I started) ...

Hides-His-Eyes · Post by **Hides-His-Eyes** » 12 Jul 2011, 22:57

300w resistor would probably be ideal wouldn't it? dissipate 50-100W without any need for heatsinks or fans, just keep the air flowing.

phatt · Post by **phatt** » 12 Jul 2011, 23:12

If you want to turn the greatest Valve Amp into a tube screamer then Yes use a 300 watt , 8 ohm resistor.

****Assuming the amp already has a great sound when cranked****
Then you need at least 3 times R for the resistive load.
A 30 Ohm Resistive load is going to deliver a far more realistic reproduction of the amps sound than 8 Ohms.

Take your signal from the load R and you will capture all of the magic.
Often those line outs are taken from preamp stages which of course misses the whole point
as it's the power tubes and transformer that you want to hear not those fizzy triodes.
Phil.

JakeAC5253 · Post by **JakeAC5253** » 12 Jul 2011, 23:34

phatt wrote:If you want to turn the greatest Valve Amp into a tube screamer then Yes use a 300 watt , 8 ohm resistor.

****Assuming the amp already has a great sound when cranked****
Then you need at least 3 times R for the resistive load.
A 30 Ohm Resistive load is going to deliver a far more realistic reproduction of the amps sound than 8 Ohms.

Take your signal from the load R and you will capture all of the magic.
Often those line outs are taken from preamp stages which of course misses the whole point
as it's the power tubes and transformer that you want to hear not those fizzy triodes.
Phil.

I am very confused by this, could you explain what you mean? You need to choose a resistive load equal to the output transformer's secondary impedance don't you? The "tonal" aspects of this design should only be applicable if and only if the output were taken from the load itself and not the preamp as we plan on doing right?. Or am I misunderstanding?

The cabinets that we run are 8Ω, so we want to set the amp at 8Ω in order to get the same sound (if it even matters, but we're playing it safe). I am just trying to choose a resistor with a wattage high enough that it will easily dissipate all heat without stress to the amp or load. You don't want to run a vintage '59 100w Marshall into a 150w or lower resistor. In the event of both your input and power signals swinging high at the exact same time, you could push peaks much higher than 100w. Like I said, just trying to play it safe.

The real catch is that we have high quality mic'd impulses of our 4x12 cabinets and speakers including power amp and mic coloration which we will run the preamp signal into in order to get useable results from. That's probably the root of the confusion, sorry I should have clarified.

Thanks for the help so far. I guess we can nix the fans then, how about the heat sink though?

phatt · Post by **phatt** » 13 Jul 2011, 00:24

Hi Jake,
If you tap the line signal from the preamp then I guess it matters little,
just hang the 300watt R load of the back and *you will at least feel safe*.

In truth it will stress the power stage far more while driving an 8 ohm load box.

A speaker is a constantly changing Z over freq bandwidth.
Depending on freq the Z load of a speaker can be anywhere from a few ohms to way above 30 Ohms.

A 30 Ohm resistive load seems to be the middle ground.

I'd much rather pull the line signal from the output transformer than rely on digital tricks but that's up to you.

What most guys miss;
triodes can do a very convincing crunch but they will never give you the compression/sag/sustain of output tubes.

Pentodes where in fact invented to overcome the severe limitations of triodes.
Also,,Triodes can't do square wave,,Pentodes CAN.
In my experience take the line from the load you make life far more rewarding.

I very much doubt that digital wizardry will be able to deliver such things.

BTW The UA (Ultimate Attenuator) uses a 30 Ohm load and I've never read of one destroying a tube amp.

I read a whole writeup on the UA and It explained the details in plain lingo.
I took that advice and I've been running 30 Ohm loads on my Amps for very long sustained periods.

ALL That comes with this simple warning;
NO matter how good you think these things are bare in mind that if your Amp is old/rare and not been serviced for a long time,
or you have never played it flat out into a speaker for extended periods of time then *you don't actually know* how reliable the rig is under extreme conditions.
So hence when the amp blows while using an attenuator most will blame the attenuator
or load box failing to realize that the Amp itself has never been load tested.

The hardest thing your amp will ever have to do is drive a load resistor.
An 8 ohm pure resistive load on an 8 ohm tap is quite streesfull and most tecks use that as a bench test.

By implimenting a higher R load will relieve some of that stress.

I can't remember where the artical write up about the UA load but google around you may find it.

Cheers, Phil.

JakeAC5253 · Post by **JakeAC5253** » 13 Jul 2011, 03:12

I did some research and I think that I would not like to run a 30Ω resistor, but I may try a 16Ω resistor as well as an 8Ω. It may sound better to run it with a higher load, but that's because you are forcing your amplifier to run at a higher efficiency. What I would be concerned about is the flyback voltages from running an 8Ω secondary into a 30Ω resistive load. I agree that speakers create a reactive load, but reactivity should only affect AC? I assume as much that the DC output still expects to see an 8Ω load regardless of phase angle fluctuations. I appreciate the idea very much, but from what I can gather the argument in favor of using such a high load is weaker than the argument against. I still may try using a 16Ω load though, split the difference sort of. Plus I can get two 16Ω resistors and wire them in parallel and then if one happens to fail the result is not catastrophic. I think that's probably the best idea.

phatt · Post by **phatt** » 13 Jul 2011, 07:25

Can I suggest doing some more research,,,, last time I looked,,, there is NO DCV on the secondary of the output transformer.

I spent the best part of 20 years trying to understand this crap and the Z load is not the problem.

If your amp can drive an 8 ohm load full pelt without issue then it will run 30 Ohms and not blink.
(I'll see if I can dig up the UA info)

There is a massive hole in the understanding of output transformers,, Z by nature is a complex equation and don't ask me to Ez-splain it all

but having spent 100's of hours reading and researching it the big problem is a lot of Valve gear runs extreme supply voltages and this is more likely the reason for failures.
Running screen grids too close to plate voltages will certainly shorten the life of any power tube.
Yes you can run the plates of a pair of EL34's at stupid voltages BUT the higher you run plate voltage the LOWER the screen has to be.

9 out of 10 Valve amps run Screen HT within a few volts of Plate. The higher it all gets the closer you come to meltdown as there is no longer any safety margin.

higher voltage just makes a couple of more watts (which can't be heard in a live gig anyway) Shortens the life of the power Valves ,,makes the power stage tighter so it sounds more like a SS Amp,,,generally does everything you don't want from a Valve Amp,,,,, but hey that is progress for you.

SE Amps will definitely have issues but a well setup push pull is hard to kill unless you are stupid.

Again I stress to you if you take a triode signal none of the above will make much difference to the tone.
If you are more than happy to record triode fizz bottles then it makes no diff. (cept your power tubes will wear out faster)

Cheers Phil.

DougH · Post by **DougH** » 13 Jul 2011, 12:29

Flyback can be an issue when you have a complex impedance that is several orders of magnitude higher than the intended output impedance load of the OT.

With a resistor you are dealing with resistance only. Resistance and impedance are 2 different things, which is what makes this tricky as Phil already mentioned. If you can find some impedance curves for typical speakers it can shed light on how much the resistance varies with frequency. As mentioned, 30 ohms resistanceis pretty much an average value over the frequency spectrum.

JakeAC5253 · Post by **JakeAC5253** » 13 Jul 2011, 18:07

Ok, I think I am getting it now. I am still having trouble with the concept of impedance used here though, it's always been something which has confused me and still does I see. Assuming we are talking about an amplifier with output secondary set to 8Ω and connected to an 8Ω speaker cabinet... when the amplifier is sitting at idle current, the load is at ~8Ω, but when the amplifier is pushing peaks the load increases to 30Ω or higher depending on frequency? Or is that wrong?

Hides-His-Eyes · Post by **Hides-His-Eyes** » 13 Jul 2011, 19:56

there is no idle current. The 'effective' resistance at any given frequency is the impedance. This is extremely variable but depends somewhat on the DC resistance (just as a giant inductor will have a large resistance?)