Aiken's Reactive Dummy Load.

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by James Freeman, Apr 27, 2012.

  1. James Freeman

    James Freeman Member

    Messages:
    374
    Joined:
    Sep 3, 2010
    So... what's inside the Suhr Reactive Load box?

    It's the impulse response era, so that's just what I do. :)
    Yet, nothing can beat a REAL tube amp, no axe, no fx, no nothing.
    It's the volume sweet spot along with the negative feedback and the presence control into a reactive load, that make the amp sound and respond to ones playing just right, and don't tell me a modeler can do that, I've tried them all.
    Besides, I can smell the Tolex cooking from the tube heat, no darn ass-fx can do that. :JAM
     
    Last edited: Feb 25, 2019
    Heph333 likes this.
  2. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    I thought I should share something I found out this weekend about mounting inductors, while finishing a reactive attenuator.

    My design uses two air-cored coils, on plastic bobbins, they are about 25mm high and 45mm diameter, for nominally 0.4 and 0.6mH. For my design I unwound them to 0.33 and 0.5mH, and I adjusted and checked them before mounting. After building, I decided to check them again since I'd just changed my meter battery, disconnecting one end of each coil. They each read about 40% higher inductance!

    What was happening was that I'd mounted them using a standard M3 plated steel bolt though the axis of the coil. The high magnetic permeability of steel compared to air added to the inductance. Once I'd removed the bolts, the values returned to where I'd set them.

    I also checked that proximity to the aluminium case was having no effect. But I'd expect that a steel case could have some influence on inductance.

    Its not enough to stop a design from working, but it does throw off all the careful maths!

    Right now, my inductors are glued in place, until I can find some non-ferrous bolts.
     
    Last edited: Feb 25, 2019
  3. James Freeman

    James Freeman Member

    Messages:
    374
    Joined:
    Sep 3, 2010
    Probably stainless steel bolts will do.
     
  4. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    Yes probably, but there are different types and many have permeabikity>1, ie, more than air. Another idea would be to use a plastic standoff, with short screws each end to anchor to case and bobbin. Or maybe just to allow for this effect in chosing the coil.
     
  5. mnemonic

    mnemonic Member

    Messages:
    155
    Joined:
    Jul 11, 2017
    When I get around to making one I was thinking how i’d attach the inductors, I think I would cut a small thin piece of wood, zip tie the inductor to that, then bolt that into the enclosure. Less permanent than glue, easier to remove or move if needed.
     
  6. James Freeman

    James Freeman Member

    Messages:
    374
    Joined:
    Sep 3, 2010
    Drill some holes and use zip ties.

    I think I used a metal bolt, but my impedance measurements were with everything assembled.
    I don't know if a bolt this size can be considered as a 'metal core' and if it actually magnetically saturates with the current that a 100W head outputs.
    I can certainly hear the 0.5mH inductor buzzing with higher volumes,, and also the output transformer is buzzing, but the tone is oh so sweet at that region.

    I think the magnetic core saturation of the output transformer is part of the tone signature of a cranked tube amplifier, those perfect oversized output transformers are lifeless, might as well use a Boss Metal-Zone into a full range PA.
     
    Last edited: Feb 26, 2019
  7. James Freeman

    James Freeman Member

    Messages:
    374
    Joined:
    Sep 3, 2010
    Here is the measurement of frequency response using very quiet pink noise into a zero-loss fx loop of my amp, (PI and Power Tubes are not clipping):

    [​IMG]


    There are high frequency filtering cap in the fx-loop or the phase-inverter, that's why you see a slope from 15kHz.
    The impedance boost is around 12db with no negative feedback.
    This speaker impedance curve along with Negative Feedback, Presence and Resonance controls will alter the amps clipping characteristics and sweet spots in various volumes,, especially when the power tubes start to clip where the nfb misbehaves.
     
    Last edited: Feb 26, 2019
  8. James Freeman

    James Freeman Member

    Messages:
    374
    Joined:
    Sep 3, 2010
    You're right @Johndh.
    There is a strong magnetic field inside the 0.5mH air-inductor when you turn the volume up,, I've held a bolt with my hand inside it and the field is quite strong.
    So yeah, remove any metallic objects from inside the inductor and zip-tie it to the case.
    Oh, and the enclosure should preferably be aluminum.
     
  9. PLX

    PLX MENSA member, Astronaut, Dated Your Mom Once Gold Supporting Member

    Messages:
    15,136
    Joined:
    Oct 24, 2017
    Location:
    Bandera, TX.
    I used zip tie mounts and zip ties for the smaller inductor.

    [​IMG]

    [​IMG]
     
    Rob s and VooDooClown like this.
  10. chur

    chur Member

    Messages:
    6
    Joined:
    Jan 22, 2017

    You arent wrong, I have smoked a Captor with my 2203 which has similar insides. Took months to get the PCB smell out of my music room as the 10watt resistors toasted the PCB and the fan blew the smell around. Fortunately the main 50watt resistors held on so no amp damage. Mind you, I've measured 100watts RMS out of the 2203 with the master on 2 and a bit.

    Since high power resistors dont appear too pricey I think I might make my own load so I can turn the amp up a little.

    It may have been covered here, but is there any opinion of whether a single high power dump resistor is better/worse than some series connected but lower power resistors which seems the norm in the torpedo etc products?
     
    PLX likes this.
  11. PLX

    PLX MENSA member, Astronaut, Dated Your Mom Once Gold Supporting Member

    Messages:
    15,136
    Joined:
    Oct 24, 2017
    Location:
    Bandera, TX.
    I started off with this whole resistive/reactive load business many years ago by using a 100W 16Ω resistor connected in parallel with a 16Ω 4x12 cabinet. It would cut the overall volume by about a third, and the resistor would get quite hot after about 15~20 minutes of playing.

    Went up and up on the wattage rating until I wound up with what you see above, which is two 33Ω 250W power resistors (in parallel) heat sink greased and bolted to a finned aluminium heat sink, inside a well ventilated case.

    I can run my 2203 with the master vol on 8 for hours like that, and the resistors barely get warm to the touch.
     
  12. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    I work out the nominal power dissipation of each resistor based on nominal amp power, and apply by a factor of x3 to spec the resistors, to allow for imperfect cooling plus extra from the amp. That is for Al clad resistors bolted with thermal grease to a substantial ventilated Al case, well drilled at top and bottom to promote convection and black-painted to promote radiation.

    Air cooled ceramic resistors, not case mounted, I want a factor of x5.

    Apart from general steady-state temperatures, there's another big advantage in case-mounted ones. There is then alot of thermal inertia and it takes 10-15 minutes or more to reach steady state with constant input. So the temp it all gets up to is an average over a considerable time, and any breaks or quieter parts reduce this value. Separate smaller resistors, even of the same total rating, heat up quicker and so can burn hotter.
     
  13. lamejohn

    lamejohn Member

    Messages:
    8
    Joined:
    Jan 29, 2019
    Updated mine a bit:

    [​IMG]
    [​IMG]

    Added a heat sink above the resistors. Since I'd used rivets to attach the resistors to the chassis I couldn't flush mount the heat sink, so I used some "heat sink plaster adhesive" to attach it, and the stuff seems to actually work pretty dang well, and the bond is, well, let's just say it ain't coming off. Got the heatsink and thermal adhesive cheap on ebay from china, so took a couple weeks to arrive.

    Removed the bolt I used running through the air core inductor. It seemed to be doing some funky things, so I used my favorite electronics adhesive (maybe my favorite adhesive period)--shoe goo. Strong hold, somewhat flexible, impervious to heat and comes off pretty easily if need be. My go to for filter caps. I glued down dense felt pads as a buffer and then glued the inductor to the pad. I also slathered the inductors in RTV silicone to cut down on the buzzing, and it seemed to help a bit. Not real pretty though...

    One other Idea I had was adding rheostats to the treble and resonance resistors for adjustability. Might try it eventually, or if I decide to build another one in the future.

    I need to get a meter that reads inductance so I can measure everything. I'm wondering if using a bunch of smaller resistors in series/parallel is adding enough inductance to throw things off. While I'm getting some good results overall, I just haven't gotten the load box + impulse response combo to sound/feel as good to me as loud ass amp + speaker cabinet. I've tried a bunch of IR's, but next step might be making my own and see what the difference is. I've tried the load box in parallel with a speaker cabinet to cut down the volume a bit, but it doesn't sound as good as just using the speaker cabinet, so I'm guessing my load box needs some testing and tweaking.

    Don't get me wrong, it doesn't sound bad, and it sounds way better than playing an amp at bedroom levels, but I'm starting to think that a pushed speaker is just part of what makes the tone and feel for me, so I'm going to do some more experimenting. Might end up building a big ass isolation cabinet eventually, despite that feeling to me like a step backward.

    And another thing I can't figure out: I'm getting what sounds like mains hum on my amp (always there with or without guitar plugged in and doesn't change volume with amp volume). I've been planning on getting a line conditioner, but the strange thing is that if I go "amp -> load box -> solid state amp -> speaker cab", and reamp the signal, it goes away, even if I crank the load box and solid state amp up it's just dead silent. Weird.
     
    Last edited: Mar 28, 2019
    mnemonic likes this.
  14. PushedGlass

    PushedGlass Member

    Messages:
    476
    Joined:
    Sep 26, 2014
    I don't think non-ferrous will get you fully in the clear, but thankfully there are nylon screws and nuts that can set you right up.
     
  15. mnemonic

    mnemonic Member

    Messages:
    155
    Joined:
    Jul 11, 2017
    Very nice! I’m still jealous I didn’t think of using a baking tray.

    Even with modellers I’ve never been able to get impulses into monitors to sound as good as a guitar cab. I eventually gave up and just use a solidstate power amp and guitar cab with my modelers.

    If you want to eliminate variables and find out if it’s the load box that’s making it sound ‘not quite right’ you can try a/b’ing your amp into a cab, and then load the amp, and reamp at the same levels into the same cab using a solidstate power amp.
     
    lamejohn likes this.
  16. lamejohn

    lamejohn Member

    Messages:
    8
    Joined:
    Jan 29, 2019
    Thanks! That's a good idea, I'll ought to approach my A/B comparisons a bit more scientifically and record them for reference if I want to make any real decisions, and I think that'll be a good place to start. I'd like to create an impulse of my cab through my amp and then A/B everything without changing the volume or mic position to see if it's right on. I was messing around a bit earlier with impulses and monitoring with headphones and made some progress using hi and low pass filtering. I think part of the problem might be some of the 'schmutz' in a guitar amp signal gets filtered out by real speakers but not by impulses.

    It's crazy how once you've got a tube amp pushing about 110db it takes all of 15 seconds to set the eq and it sounds amazing. But trying to get a good sound at sane levels I spend more time fiddling knobs than actually playing.
     
    Last edited: Mar 29, 2019
    mnemonic likes this.
  17. mnemonic

    mnemonic Member

    Messages:
    155
    Joined:
    Jul 11, 2017

    Just saw this posted on another part of the forum (gut shots thread in amp forum).

    Looks like the Aiken circuit.

     
    PLX likes this.
  18. PLX

    PLX MENSA member, Astronaut, Dated Your Mom Once Gold Supporting Member

    Messages:
    15,136
    Joined:
    Oct 24, 2017
    Location:
    Bandera, TX.
    Thanks for posting this.

    Been curious to see the guts of one of those. :aok
     
  19. Husky

    Husky Gold Supporting Member

    Messages:
    10,143
    Joined:
    Nov 5, 2008
    It has inductors, resistors and capacitors which is what is required to make a proper curve that truly follows a speaker/cab. No way around that. Randall is a smart man and a friend, he knows what is required. Jim Kelley worked on mine, he also knows.
     
    Last edited: Apr 7, 2019
    GaryMcT, Tony Bones and mnemonic like this.
  20. Tony Bones

    Tony Bones Member

    Messages:
    1,116
    Joined:
    Nov 1, 2014
    Location:
    beneath the moon
    Exactly. Randall Aiken did a great job showing us what goes into a speaker emulator and even detailing the specific requirements of each of the components. As I said, he did a great job, but I don't think we can name the circuit after him.

    At the end of the day, you need a series LR element to get the rising impedance with frequency, and a parallel LCR element to mimic the speaker resonance. I don't know of any other way to do it, so any load box with similar frequency response is going to have those electrical components.

    But thanks again to Randall Aiken for showing us how to DIY one ourselves.
     
    reaiken likes this.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice