DIY Loadbox (and/or DI) questions

Tube or solid-state, this section goes to eleven!
Post Reply
User avatar
MentalDekay
Information
Posts: 2
Joined: 21 May 2019, 15:20
Has thanked: 7 times

Post by MentalDekay »

Long time stalker, first time poster..
(Mods: feel free to move this post to the most appropriate sub-forum if this one is wrong.)

I've been wanting to have a means of hooking my Vox AC4 directly into my Behringer UMC202HD for computer recording. This would be to bypass the amp's speaker. Obviously, I shouldn't just send the 1/4" cabinet output to the AI as that would likely melt something. The answer seems to be "loadbox", but all the commercially available ones seem extremely expensive given what they're actually doing (tapping the characteristics of the speaker drive signal to a safe, line level). I tried searching the forum, but when dealing with stomp box electronics the words "load", "tube", etc are very common.

Also, I apologize for some of my ignorances, I'm a EE by training but I don't work with tubes, audio signals or speakers (I design high speed serial buses and automotive computing platforms). I know that the load of a speaker is not constant over frequency because of many things (magnetics/inductor/transformer, cone and enclosure characteristics, etc).

So, for those that have made their own loadboxes:
  • How much difference is there in using a (almost purely) resistive load from as compared to an inductor + resistor load?
  • How do you size the inductor to handle the power from the tube amp?
  • Can you resistively divide down the signal then add a small inductor in parallel?
  • Are there any other considerations to allow flexibility between my little 4W class A and a ~50W amp? Obviously matching the speaker/cab impedance, but what else?
  • Would it be wisest to cut the signal power down to line levels or below then redrive them with a non-distorting opamp circuit?
Thanks for any tips!!

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

IMO, the best way to this is the combination of a reactive load and a speaker simulation.
The reactive load simulates the speaker impedance depend on the signal frequency
,
the speaker simulation the frequency response of the speaker.
If the reaktive load rated up to 50 Watt then it works for Amp with lower Wattage too.

User avatar
phatt
Transistor Tuner
Information
Posts: 1224
Joined: 04 Aug 2010, 05:17
Location: Morayfield SE Qld AU
Has thanked: 8 times
Been thanked: 343 times

Post by phatt »

You can spend a fortune and drive yourself nuts with big complex ideas or you can simply use this.
I built this for live use, driven by a small 10 Watt valve amp and reamped into a much bigger Amp. I used this system for about 4 years.
I can't see why it would not work for recording as most of all the other fancy tricks can be done with software now. [smilie=a_whyme.gif]
I'm busy now but if you want an in depth explanation I'll write it up for you.
Meantime if you want to hear it working into another amp here is a Vid i made up a while back. It's not about load boxes as such but I'm using it to do this vid to keep the noise down as I live close to neighbors.
The only thing between the 2 amps in the vid is the load box.

Phil.
Attachments
SPEAKER2LINE.JPG
SPEAKER2LINE.JPG (26.28 KiB) Viewed 3528 times

User avatar
vtguedes
Information
Posts: 22
Joined: 04 Oct 2017, 22:07
Has thanked: 5 times
Been thanked: 6 times

Post by vtguedes »

Hi,

My passive reactive loadbox is an 8 ohm 8 inch 30W speaker with its cone removed. I use it to record silently using a Fender Champ clone (6 watts-ish).

To plug into the audio interface (i use an UMC202 just like you), I have used it with a 10k resistor and a 5k potentiometer associated in series, then both are connected in paralel with the load. This also works when connecting to the speaker instead of the load.

Here is a really simplified schematic of what I have done.
SCHEMATIC.JPG
You will want to use a DAW software with some impulse responses to simmulate the speaker cabinet and mics. Also be aware that the speaker with the cone removed won't move air (as there is no cone) but it will vibrate a lot.

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

My passive reactive loadbox is an 8 ohm 8 inch 30W speaker with its cone removed.
A speaker with a removed cone is not really a reactive load, it is only an inductance.
The coil shows not the desired impedance response of the complete speaker respectively of the box in which it is installed.
This is due to the mechanical and acoustic feedback to the elektrical side, with both are transformed to electrical parameters.
Have a look at the equivalent circuit diagram:
Mechanical-analogous-circuit-for-a-dynamic-loudspeaker-in-infinite-baffle.ppm.png

User avatar
okgb
Diode Debunker
Information
Posts: 768
Joined: 16 Mar 2010, 03:58
Has thanked: 226 times
Been thanked: 38 times

Post by okgb »

Has anyone done extensive testing to compare purely resistive loads Vs ones with inductors ?
I did some at Winnipeg studio Groove jungle with Howard Klopak , recording a marshall head loaded both ways then into an ada speaker sim and
it was tough to tell, there didn't seem to be much difference, this was as a load only and not attenuator into a speaker, so it didn't seem to influence
the sound of the transformers contribution when reamping
I never thought the old Rockman power soak was so bad!

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

I never thought the old Rockman power soak was so bad!
I built such a similar power soak using cascaded resistor pi-attenuator circuits in the late 1970th,
but without a rotary switch.
I had got output jacks in between whereby the following part of the cascade had been switched off by the normaly open contact of each jack.
I built the device for a friend of mine and installed in an old VOX AC30.
It works fine, but I cannot remember if it affected the sound, but there were no complaints.
At that time I have should patent it, but it would therefore have been an illusion because patenting costs too much. :(

If we talk about simulation of an ommited speaker for recording a reactive load is needed.
The impedance of reconed speaker rises nearly linear with the frequency,
nearly because the inductance of the coil changes while the coil is moving.
In the impedance response he resonance hump is missing and the high frequency behaviour is different.

Here a article on reactive load:
reactivLoad.pdf
(101.53 KiB) Downloaded 198 times

User avatar
vtguedes
Information
Posts: 22
Joined: 04 Oct 2017, 22:07
Has thanked: 5 times
Been thanked: 6 times

Post by vtguedes »

Manfred wrote:
My passive reactive loadbox is an 8 ohm 8 inch 30W speaker with its cone removed.
A speaker with a removed cone is not really a reactive load, it is only an inductance.
The coil shows not the desired impedance response of the complete speaker respectively of the box in which it is installed.
This is due to the mechanical and acoustic feedback to the elektrical side, with both are transformed to electrical parameters.
Have a look at the equivalent circuit diagram:
Mechanical-analogous-circuit-for-a-dynamic-loudspeaker-in-infinite-baffle.ppm.png
Thanks for the info! Didn't think so many stuff would change the impedance the output transformer can "see". I guess there is still a lot for me to learn.

Anyway, my load box is really simple, works and sounds good with the soundcard when using some celestion impulse responses, i guess that is what OP wanted.

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

hi vtguedes,

did you remove the cone and the suspension of the speaker or only the cone of it?

User avatar
vtguedes
Information
Posts: 22
Joined: 04 Oct 2017, 22:07
Has thanked: 5 times
Been thanked: 6 times

Post by vtguedes »

Hi Manfred,

Sorry I didn't really explain what I did to the speaker to use it as a passive load.

I removed just the cone and secured everyhing else with a bunch of zip ties to "simulate" the cone tension/resistance and keep everything in place.

User avatar
Manfred
Tube Twister
Information
Posts: 1937
Joined: 04 Apr 2009, 23:42
Has thanked: 1671 times
Been thanked: 1344 times

Post by Manfred »

Thanks, I see, so you have achieved an approximation of the impedance characteristic in this manner.
That was a good idea, have you took a shot of it? :thumbsup

User avatar
vtguedes
Information
Posts: 22
Joined: 04 Oct 2017, 22:07
Has thanked: 5 times
Been thanked: 6 times

Post by vtguedes »

The photos from the thing got lost with some old computer aprox 15 years ago.

The box is on my parent's house but they don't live far from me, when I get the chance I'll take a new one or might make a new load box since I have a spare 6,5" speaker from an old practice amp.

User avatar
phatt
Transistor Tuner
Information
Posts: 1224
Joined: 04 Aug 2010, 05:17
Location: Morayfield SE Qld AU
Has thanked: 8 times
Been thanked: 343 times

Post by phatt »

okgb wrote:Has anyone done extensive testing to compare purely resistive loads Vs ones with inductors ?
I did some at Winnipeg studio Groove jungle with Howard Klopak , recording a marshall head loaded both ways then into an ada speaker sim and
it was tough to tell, there didn't seem to be much difference, this was as a load only and not attenuator into a speaker, so it didn't seem to influence
the sound of the transformers contribution when reamping
I never thought the old Rockman power soak was so bad!
Yep hold that thought and don't place too much expectation on all the maths.
I have no doubt that some highly qualified folk who work out those complex equations are correct and effect the result but have you noticed that it's never mentioned that the end result may only make a 1% difference to the sonic result.
Jezz,, most of the time it's for heavy distortion guitar sounds anyway so it's hardly worth a 4 year stint at Uni just to understand what the hell it all does if in the end it's hard to tell any difference. [smilie=a_whyme.gif]

I went in circles for some time and then I read an article explaining the workings of the *Ultimate Attenuator*.

As I'm the kind of chap who is always looking for the one line that gives you an Arrh huh moment I found this eye opening and a far better & simpler explanation of what happens.
The line read;
*The 30 Ohm load resistor in the UA is an Average Z over the whole bandwidth for an 8 Ohm tap* ARRH HUH of course!!! [smilie=a_bravo.gif]

I walked straight out to the shed and changed my 8 Ohm resistor for 24 Ohm and the that changed everything,, Bingo close enough for me. :horsey:
That changed My little 10 Watt Valve amp from and over compressed tube screamer sound to a real OD monster sound at any level I needed. Never bothered anymore with all the other stuff,, just too hard and likely little benefit to be had.

Another point you might like to ponder;
When you read all those complex articles and words like Resonance and Inductance are explained in greek. :scratch:
Stop and ask yourself, What does it do to the sound? Basically it changes the TONE.
Hum lets see....well why not use a tone control circuit? :secret:
yes I know it won't be perfect for the experts but will give you a lot of variation.

I built a load box for a chap years back, a little 10 Watter into my load box (as posted earlier) into a Tandy HiFi Graphic EQ >into a bigger wattage SS combo.
That was at least 10 years back and 5 years back he was still using it. :thumbsup
If you want even more control try some Para EQ circuits as they can do deeper curves/notches than a Graphic EQ.

Keep in mind that no matter what you use there will never be perfect results as what you hear from a real speaker at full volume is reflecting off surrounding environments which is a whole other subject. [smilie=poke.gif] :|
So when you plug in on Saturday and it sounds good then Sunday it sounds different. Why,,Because Saturday was sunny but Sunday was cold and wet.
Yes even the weather can effect the sonic result. :hmmm:

Hope it helps save someone out there from all the hassle of wasting time on expensive gear. :thumbsup
Phil.

User avatar
okgb
Diode Debunker
Information
Posts: 768
Joined: 16 Mar 2010, 03:58
Has thanked: 226 times
Been thanked: 38 times

Post by okgb »

Thats a great reply , thanks! Do you happen to know where the Ultimate attenuator article that you read is ?

a couple companies have a great angle pitching their re-amps to make a small amp into a loud amp, also adding fx loops
to amps without them, and fryette upping the game to " real " tube power re-amp, ohhhh , bad cat makes a S.S. ver too

User avatar
phatt
Transistor Tuner
Information
Posts: 1224
Joined: 04 Aug 2010, 05:17
Location: Morayfield SE Qld AU
Has thanked: 8 times
Been thanked: 343 times

Post by phatt »

Jezz a big ask as that was close on 20 years back,, lol.
This is not the article I read but gives the same info;
https://legendarytones.com/the-ultimate-attenuator

As like this article notes I found little need for treble or bass boosting. In my case I'm Not using it as an attenuator only using the load resistor and a voltage divider to tap the line out. The 30R resistor is the important part in either case as it lets the Amp open up much better.
Any extra tone shaping is done post load box and I assume when recording direct there would be need to add final tone shaping.
(I doubt the value of resistive load has to be exact and with big rigs you may want to be careful going to large ,,especially if they run EL34's.. a lot of EL34 Amps run EL34's way past the Voltage limit of the screen grid which is a common failure mode :twisted: )

Look up the Guytron GT100 amp which is two amps in one head but the principle is basically the same trick.
A small Amp driving a Resistive load and tapped into a much larger Amp, It's all Valve but way to heavy for me as well as way to expensive so I opted for the small Valve head> load box and second amp is SS. My main second amp is an old Laney keyboard amp which has only basic bass treble volume and a spring reverb. The Vid link I posted earlier used the SS Laney as the second Amp.
(Bear in mind I'm not a pro player so it's all on a strict budget for me)

EVH used a load box after a Marshall and the second Amp/'s was all SS,, and few would argue it lacked the Valve sound. [smilie=a_whyme.gif]
So if SS was good enough for Eddie then why try to reinvent it with valves for the 1 or 2% improvement??? :blackeye

There were some sound clips on the original Guytron site years ago which sounded fantastic but that has all changed now but I think there are a few Utube clips showing off the amp.

The difference between EVH rig and Guytron is likely due to the fact that Eddie's rig would have been Fixed bias whereas the first amp in the Guytron is Cathode biased and they tend to have a sweeter softer transition into OD than most Fixed biased amps ever dream of.
The power stage in my Vid is Cathode biased and it responds way better than all the other fixed biased rigs I've messed with. :secret:

I did look into Kevin O'Conner's power scaling but the Guytron sold me,,, then not long after that the 30 Ohm load trick from the UA gave me the missing link to the magic Cigar. :horsey:
cheers Phil.

User avatar
MentalDekay
Information
Posts: 2
Joined: 21 May 2019, 15:20
Has thanked: 7 times

Post by MentalDekay »

Wanna say thanks for all the inputs.

I think I'll start off with a purely resistive load. I'm going to eventually sweeten it with selectors for 4/8/16 ohms and maybe look into some HPFs and LPFs to round the highs and lows (to sorta emulate a speaker freq response). I also liked the option to bypass the loadbox when connected to a amp/speaker/cab.

Feel free to carry on the conversations, a lot of knowledge has been dropped!

When I get something physical built, I'll try to post an update.

Thanks!
-des

Post Reply