Aiken's Reactive Dummy Load.

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

  1. TomOlsen1

    TomOlsen1 Member

    Messages:
    41
    Joined:
    Mar 25, 2008
    Location:
    Johannesburg, SA
    Thanks for the explanation. It's helped me understand what I need to change with my existing DIY attenuator/loadbox. In my case, I'm already using a 4 pole 3 position rotary to switch in different values, but only 3 of the 4 poles, which may give me some options.

    Noted regarding the line out.

    How hot does your unit run? I've messed around with adding a speaker driven fan using the powerbrake implementation as below. But I didn't much like what it did to the sound - mind you, I was connecting it across the lugs on the input jack, perhaps this isn't the right way to do it - If I look at the 2nd schematic the fan comes after a 56R resistor - (2nd schematic has an error in the bridge rectifier wiring)

    The one I used to build the fan supply.
    [​IMG]
    More detailed schematic, but with error in bridge rectifier wiring
    [​IMG]

    Has anyone else added a fan in this manner - any suggestions?
     
    Last edited: Mar 12, 2018
    guitarobert likes this.
  2. PLX

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

    Messages:
    15,136
    Joined:
    Oct 24, 2017
    Location:
    Bandera, TX.
    The one I constructed has a case fan, but it is powered from a separate wall-wart type power supply.

    Didn't want a fan in the circuit at the risk of noise.

    [​IMG]

    [​IMG]
     
    Last edited: Mar 13, 2018
    Roe and TomOlsen1 like this.
  3. TomOlsen1

    TomOlsen1 Member

    Messages:
    41
    Joined:
    Mar 25, 2008
    Location:
    Johannesburg, SA
    Ha! I've actually saved those pics as a reference before I posted, what a great build.

    Making this post from my phone, so please excuse lack of pictures for now, will update further once back to my workbench.

    I was just on a business trip to Germany, so I picked up one of these Thomann PA-100 attenuators nice and cheap for the purpose of reverse engineering the fan and as a donor for this project.

    https://m.thomann.de/gb/harley_benton_pa_100_power_attenuator.htm

    They're basically the Jet City Jettenuator design (schematic on FSB) but with the addition of a fan and overload led. I got the chance to take it apart last night and see how they'd implemented the fan, then use remaining space in the chassis to add the reactive bits if possible.

    [​IMG]

    It's essentially doing the same thing as Badside's unit without the reactive component... -6db pad using 50w alu resistors prior to a variable L-pad - although the pad in my unit isn't quite keeping 15 ohm load throughout sweep, it goes as low as 12 ohm in some spots.

    The fan component is pretty much the same as the 2nd powerbrake schem above, but with the addition of a voltage regulator and they've dropped the 56R ceramic preceding the bridge rectifier down to 15R. I tested the concept on my own DIY L-Pad last night using the powerbrake schem with ceramic resistors of 15-68R (my thinking was that in the marshall they are using a resistor of 5x the load, in the Thomann same as max input since that one goes from 4-16 ohm) tried it at input and output positions, as well as in bypass, -6, -9 and load settings. Settled on connection at input jack and 15R ceramic for now, seems to get the fan working at band volumes.

    [EDIT: Eventually changed the 15R to 56R as used in the power brake.]
    [EDIT #2: 56R was producing 26vdc to the fan supply when hooked up to my cranked JMP - adjusted resistor to 75R by using a 120R and 220R in parallel, now max voltage to fan supply sitting around 23vdc.]

    Very happy with the result. The negative impact on the sound I was getting without the resistor seems to be gone. I'll knock out some clips to be able to compare the sound with and without the fan connected, but last night I couldn't tell the difference in person.

    Now I just need my reactive parts to put in between the shunt resistor and connection to ground. (And I might beef up the 50w alu to 100w while it's opened up...)

    [EDIT: looking at the jettenuator schem above, it's missing the switching jack on the 16 ohm input - for a better reference the schem from FSB should be used...]
     
    Last edited: Mar 20, 2018
    guitarobert and PLX like this.
  4. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    This is a great and insightful thread with some very smart contributors. So I’m hoping that I could get some more thoughtful discussion about some thinking on resistive/reactive attenuators. Particularly, whether it is possible to ‘tune’ the tone of a purely resistive attenuator, to get close to the response of a reactive load. Don’t laugh yet… (and I hope that this is near enough to being on-topic)

    Resistive attenuator designs

    I’ve been building and testing quite a few purely resistive attenuators, for use on my Marshall Vintage Modern 2266c. This is a 50W amp with a pair of 12” Greenback speakers for a nominal 8 Ohm speaker load. Its partly because I want one for use, and partly to learn by testing, and wishing to start simple.

    The attenuators are very basic, just a single attenuation value with no switching. Typically, there are three resistors in a ‘star’ configuration. This allows me to pick values to target attenuation db, plus both input and output impedance. I maintain the nominal input impedance as seen by the amp at close to 8 ohms, including the speakers. I’ve most recently been targeting around -12db basic attenuation, bringing a 40W amp down to 2.5W for home use.

    Here are three resistor layouts that all do -12db with around 8 Ohms seen by the amp.

    [​IMG]

    Type A on the left, is a simple L-pad and it sounds noticeably dull. Type B in the middle sounds quite good, its clearer with much nearer unattenuated tones. Type C gets brighter again, see analysis below. The main difference between them is the output impedance as seen by the speaker. Based on nominal resistance values, speaker in Type A sees 2 Ohms, B sees 6.6 Ohms and C sees about 26 Ohms. Type C is an extreme example, I tested it and it is a bit bright and harsh, but A and B are quite feasible.

    I’ve built Type B into a box and have been using it all volume settings. Comparison of recordings attenuated and not attenuated, using a looped clip then miced and normalised for equal volume show negligible audible tone difference (though a small response difference is measurable).

    So what puzzled me is that I'm getting this good result by having quite a high output impedance from the attenuator, so far as I understand, 6.6 Ohms is much greater than most amps output Z.

    Amp with reactive load: response

    Aitkens page below shows how the electrical response of an amp varies with frequency when driving a reactive load.

    http://www.aikenamps.com/index.php/parallel-attenuator-loads

    This could be a real speaker, or a load box as in this thread. His page is focusing on how this changes if a resistive load is in parallel, but Im just looking at the full response curves in green. For amps with or without feedback loops, the response varies in a similar fashion to that expected from the speakers reactive impedance. For an amp with a feedback loop (such as mine), there is a lift in response of around 3db from 500hz up to 5khz due to speaker inductance, and a boost in bass due to resonance, as shown here:

    [​IMG]

    Clearly the real amp and reactive speaker load are interacting intimately to make this response.

    With a resistive attenuators at -12db, amp and reactive speaker load are almost completely separated. The amp is seeing almost a totally resistive load and the speaker sees just a resistive output Z. About 15/16ths of the amp power is going into resistors in this case.

    But Spice analysis of this situation shows how the reactive impedance of the speaker could interact with a pure resistive output Z to get very close to unattenuated frequency response, at an attenuated output level.

    I used the reactive load box design from Aitken, scaled to an 8 Ohm version, to represent a real speaker, and tested the calculated frequency response, in terms of signal seen by the load, within the range of output impedances from 2 to 26 Ohms, as for designs A to C above:

    [​IMG]


    Note: in the Spice program I use, I can’t vary more than 2 components at once, so the plots below show the signal seen by the speaker load, with varying amp output impedance, but without controlling overall attenuation level – but it’s the relative shapes vs output Z that are important.

    [​IMG]

    The plots show the signal seen by the speaker, with output Z from the attenuator varying from 2 to 26 Ohms.

    The arrangement Type B that I built, with an output impedance of 6.6 Ohms, is very close to the red curve on that graph, but at -12db. It shows just under a 3db rise in treble from 500Hz to 5kHz, ie very close to the full amp/reactive load curve from Aitken. I reckon that helps to explain why it is sounding pretty good!

    As the output Z increases further, about 3.5 more db of high and low relative boost is available.

    This seems like a helpful idea, ie controlling Z to make simple attenuators sound more like an unattenuated amp at least in terms of basic frequency response. So what’s missing? I can’t see any undue risks, since resistor attenuators can be built very robustly and are fine for loading a valve amp. At moderate distortion levels, I think the match is reasonably valid, but at high power-amp distortion, there could be tonal differences. The tubes will still be saturating as they drive into the resistive load, and this will come through in the tone, but Id guess that some of the more subtle interactions of tube/OT/real speaker will be suppressed by the resistances between them.

    -12db Attenuator build

    This is what I built, based on the Type B design. its all inside a 120x90mm aluminium box, with raised feet and well drilled top and bottom for ventilation. The 5 ohm resistor is 2x10 in parallel, and those and the grounded resistor each dissipate around 12W based on a 40W amp output. They are oversized at 100W rating. The output series resistor is a 10W wire-wound, but it only dissipates 1.25W, so its all overbuilt.

    [​IMG]

    [​IMG]


    That's where I've got to. I’d be very grateful for any comments or insights.
     
    Last edited: Apr 28, 2018
    davidespinosa and Shiny_Beast like this.
  5. pdf64

    pdf64 Member

    Messages:
    5,783
    Joined:
    Aug 9, 2008
    Location:
    Staffordshire, UK.
    Where have you got the 2 ohm amp output impedance from (sorry, I haven't read the thread)?
    A tube guitar amp's output impedance can't really be characterised as being one value.
    At high signal levels, open loop, generally it may be around 2-4 x the nominal intended load impedance.
    With the normal level of global NFB, that may about halve, though presence / resonance controls in the loop will move things towards open loop in their frequency ranges, as they are advanced.
    At low signal levels, the output impedance may tend to be higher.

    It may be worth bearing in mind that I think one of the goals of the reactive load is for the amp to have a similar response to that which it would have to a real cab; their response into a resistive load is somewhat different.
    But given that your approach can't achieve that, I think you're probably heading in the right direction, ie in avoiding damping the speaker too much, in order to get a more realistic tonal balance from the speaker.
     
  6. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    Thanks for your comments. I've seen several references to tube-amp output impedances in the general range 1 to about 4-5 ohms. Here's one:

    http://www.diyaudio.com/forums/tubes-valves/165634-typical-output-impedance-tube-amps.html

    2 ohms was just a value in that range

    Also, Here's Aiken running a design example on a guitar-amp output stage, ending up with 5.2Ohms with NFB:

    http://www.aikenamps.com/index.php/designing-for-global-negative-feedback

    But higher values of Output Z would make some sense, and they definitely change with conditions and frequency. It also helps to explain the measured output plots with a reactive load.

    But for what I'm looking at in my post above, the amp output Z doesn't make too much difference since the speaker doesn't really see it through all the resistors. But I can follow a real amp/speaker output frequency response (ie as Aiken plotted) reasonably well by tuning the attenuator output resistance to work with the reactive speaker impedance. At least, it is much better than a typical two-part L-pad, which leads to a very low output impedance at high attenuation.

    Also, definitely not knocking the reactive load-box with slave amp concept - its a high level of performance than what I'm doing, but at considerable more cost and size.
     
    pdf64 likes this.
  7. pdf64

    pdf64 Member

    Messages:
    5,783
    Joined:
    Aug 9, 2008
    Location:
    Staffordshire, UK.
    The context of that link is hifi, the objectives there being linearity, low distortion. Their expensive, high spec transformers allow lots of global NFB, and local NFB techniques such as UL, cathode windings etc reduce output impedance further.
    You may appreciate that tube guitar amps are a rather different, cruder, simpler thing.

    Note that his starting point is that he assumes that the open loop output impedance ('internal output impedance') is equal to the nominal load impedance, 16 ohms in this case.
    My finding is that actual measured, high signal, open loop output impedance may be rather higher than that.
    Even more so with your amp, which I think uses KT66 beam tetrodes, rather than the EL34 pentodes in Aiken's 'typical' amp.
     
    Last edited: Apr 28, 2018
    Johndh likes this.
  8. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    Thanks very much, that's the kind of info I was hoping to learn.

    I think you are right. Going back to Aikens plots where he added an 8Ohm resistor as half of the load, it is adding damping that apparently reduces the bass resonance by a significant factor, much more than if the amp was already damping the speaker with just a couple of Ohms Z. The speaker is feeling a much greater change in damping than that would imply, suggesting a higher output impedance from the amp.

    I tried my Spice runs again, this time with a very high Amp output Z (arbitrarily 50 Ohms just to try it - but I know its a much more complex impedance than that). For my -12db resistor networks, all the frequency response shapes stayed the same with about a 0.25db extra calculated boost at high and low frequencies.
     
    pdf64 likes this.
  9. nonost

    nonost Member

    Messages:
    1
    Joined:
    Jul 4, 2017
    Well, this is a great thread. And the soundclips posted sound pretty damn good. Does anyone know where to find the inductors in Europe?

    I'd love to build a 30w version of it.
     
  10. Mike Lind

    Mike Lind Member

    Messages:
    49
    Joined:
    Jun 13, 2017
  11. mnemonic

    mnemonic Member

    Messages:
    155
    Joined:
    Jul 11, 2017
    late pass on my post, but I've read through this thread today since over the last week I've picked up an interest in attenuators and reactive loads, figuring out how they work, etc. Just want to say, thank you so much for everyone in this thread, real wealth of practical information.

    Anyway, I got ahold of the schematic and layout for this tubetown reactive load from someone who built one (they enjoy it and say it sounds great), it is indeed based on the Aiken schematic but with different values.

    [​IMG]

    [​IMG]

    I kind of want to build my own reactive load box and I'm trying to size up if its a better idea to buy the parts from erse audio that James Freeman used, or just buy this kit.

    So I've stumbled my way though learning how to use LTspice this evening, and after some trial and error I've managed to plot out the impedence of this schematic:

    [​IMG]

    I figured I must have done something wrong since that resonant peak is redonkulous in size. If I lower the value of the resonance resistor(s) to about 50R, it looks more normal, but I couldn't figure out why they used 235R (two 470R in parallel) in this position. Then I remembered what Randall Aiken said about the DC resistance of the inductors and how it must be very low (under 1 ohm) or the reactive element is defeated.

    I looked up the inductor used here http://www.mundorf.com/en/?category=pro&menu=coils&content=mcoil_l_pro

    and the 27mH inductor called for in the schematic has a DC resistance of 6.44R according to this chart. I plugged that series resistance into LTspice and the impedence plot suddenly looks a lot more normal:

    [​IMG]

    the treble slope did not change, it just looks bigger since the scale of the graph changed due to the smaller low resonant peak.

    My question is: Randall Aiken was adamant that you must use inductors with a very low DC resistance, in the real world, can a large inductor with a high DC resistance as above be compensated for by using a much larger resonance resistor?

    I don't know enough about this kind of stuff to determine if its something that works well in simulation but not in reality. I haven't heard any complaints about this reactive load box (though I've seen very few reviews) so maybe its fine? I guess it would make building this project more cost-effective since we wouldn't necessarily need to find very low resistance inductors.

    For what its worth, I only learned how to use LTspice in the last hour, so theres a non-zero probability that I'm doing everything wrong.

    I plugged James Freeman's final values as on post 1 of this thread into LTspice and I got this plot:

    [​IMG]

    which seems to mostly line up with his 'final' plot vs. his cab, though that resonant peak is higher in my plot. I did also include the DCR and amperage rating from the ERSE website for the inductors, so not sure if that has an effect on the simulation.


    Side note, this program is pretty fun.
     
    Heph333 likes this.
  12. PLX

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

    Messages:
    15,136
    Joined:
    Oct 24, 2017
    Location:
    Bandera, TX.
    FWIW, I used the ERSE Super Q 20mH for the reactive Inductor.

    I just measured the DC resistance of it and got 0.7 Ohms.


    cheers!
     
    mnemonic likes this.
  13. Johndh

    Johndh Member

    Messages:
    341
    Joined:
    Sep 29, 2017
    Hello, Im interested in these too. But what do you really want to do? Do you need a reactive load box from which to take a signal for IR or slave amps? Or do you just want an attenuator between your amp and speaker to reduce volume? If the latter, there are simpler ways based on rhe right resistive design. Ive been building these over the last year, gradually improving, and several other guys have built them too. Also with an optional reactive module. My journey is here, and you might be interested to take a look: (current best design on p7, the first few pages are my learning curve))

    http://www.marshallforum.com/threads/simple-attenuators-design-and-testing.98285/
     
    Heph333 and mnemonic like this.
  14. mnemonic

    mnemonic Member

    Messages:
    155
    Joined:
    Jul 11, 2017
    Very interesting thread, thanks for posting that!

    Personally, I’m only interested in knocking a few dB off my recto, since the volume where the fizz goes away is a bit painful in my room. I still want to be in the 95dB to 100dB range of volume. Just want to get the to ‘sweet spot’ a bit earlier.

    I think a reactive load box + reamp is likely overkill for this situation, but it’s still a very interesting topic. It’s something I want to try, if only to satisfy my curiosity.

    Given the simplicity of the attenuator designs in that thread, I may try banging together the 16 ohm version. I’ll be ordering parts soon for some other projects, can’t think of a reason not to!

    I also recently ordered a used thd hotplate off eBay so it will be interesting to compare them when it shows up. I am a bit pessimistic about resistive-only loads given I mostly use my recto in modern mode, where the negative feedback loop in the poweramp is disconnected. But it was cheap enough I can resell for probably the same price.
     
    Johndh likes this.
  15. JimSoprano

    JimSoprano Member

    Messages:
    265
    Joined:
    Feb 28, 2015
    Location:
    Amsterdam
    Anyone has any experience with Aiken design vs. Torpedo Captor, Suhr RL, etc?
     
  16. PLX

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

    Messages:
    15,136
    Joined:
    Oct 24, 2017
    Location:
    Bandera, TX.
    Here's a gutshot of a TwoNotes Torpedo :

    [​IMG]

    Good luck cranking up a 2203 into that and not starting a fire. :D
     
  17. Husky

    Husky Gold Supporting Member

    Messages:
    10,143
    Joined:
    Nov 5, 2008
    I’m sure it can handle it but the measured impedance curve from this and some others looks more like a smiley face rather than a speaker in a cab impedance curve. How much that matters to you is a decision for the consumer. I believe in one of Pete Thorn’s videos he shows the tonal difference between different loads using the same IR, the differences are not small and that ignores the feel.
    We also just released this video to help people get up and running.
     
    Last edited: Nov 2, 2018
  18. lamejohn

    lamejohn Member

    Messages:
    8
    Joined:
    Jan 29, 2019
    Hey, signed up to say thanks to everyone for putting in the work on the design of this thing and making it freely available. Just got done with mine so thought I'd post it here:

    [​IMG]
    [​IMG]
    [​IMG]

    ******EDIT***** I used 5w resistors for the 50 ohm treble and 75 ohm resonance, as my ltspice models never showed them taking much more than a few mw, however reality seems to be a bit different and I just fried the treble resistor doing a test, so time to get the 50 watters like everybody else used :jo******

    Stuck to the design in the first post of this thread, except I added a 4 OHM/8 OHM switch (most of my Fender amps are 4 OHM). ERSE didn't have the 200uf capacitors listed so I got two 100uf and put them in parallel. Went with multiple 50w 4 OHM resistors instead of a single large one, mainly because mouser had them for about $2.70 apiece when buying 10. Slammed this thing for about an hour with a triple rectifier (150w), and it certainly got hot, but nowhere close to the 200*c max. Measured about 45-50v coming out the triple rec when cranked to the max! Used an old cake pan for the chassis. Was probably right about $200 for all parts and shipping. Thanks again to everyone for posting your designs and doing the testing, this thing is awesome!
     
    Last edited: Jan 30, 2019
    mnemonic likes this.
  19. mnemonic

    mnemonic Member

    Messages:
    155
    Joined:
    Jul 11, 2017
    Looks good! How well does it work with the triple rectifier?

    I’ve been planning one of these for a while but still haven’t got around to it, I think I will start ordering parts this month though.

    Good use of a cake pan, I’ve been trying to think what I’d use for an enclosure.

    I’ll predominantly be using it with my Dual Rectifier which is why I ask. It’s sweet spot is just a bit too loud for bedroom playing.

    I typically use Modern mode, which disconnects the negative feedback loop in the poweramp, and this seems to make the sound much more sensative to the load.
     
  20. lamejohn

    lamejohn Member

    Messages:
    8
    Joined:
    Jan 29, 2019
    Thanks! I haven't had a lot of time to try out the recto with it yet (mine's a late 90's 2 channel, so I'm not sure if yours is 3 ch/multi watt might be different?) So I've only got an initial impression. Tried a bit of reamping with a solid state QSC power amp and also used impulse responses on the computer with headphones (i'm new to that so have a lot to learn and have only tried a couple of the free ones). Also, since the mesa's have a 'slave out' I tried reamping that signal vs the line out of the load box---and I'm kinda undecided so far about what works/sounds best so far. (too many options!) My main use will be for silent recording and jamming, but it kinda seemed that a cranked recto reamped at low volume sounded pretty similar to a recto played at lower volume (for the most part), though when your really get the power amp going it does really change the character of the amp, and it's not something I'd ever get to fool with without this load box. Lots of possibilities!

    Also, to anyone who has built one, if I do a continuity test between chassis ground and input (+), I get a beep, and I'm curious if this means I've got a short somewhere or if since it's a low resistance circuit with the inductors that my meter reads it as a short (my fluke reads continuity <70 OHms), so I think that's the case, as I can't find anything out of place in my circuit. Thanks!
     
    Last edited: Jan 30, 2019
    mnemonic 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