1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

Solid State Spring Tank Reverb Circuit

Discussion in 'Amp Tech Center' started by robrob, Feb 24, 2016.

  1. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,260
    Joined:
    Dec 29, 2012
    Location:
    United States
    A few days ago printer2 pm'd me the address of an interesting web page that pointed out how simple the Fender Hot Rod Deluxe's solid state spring reverb was.

    I know very little about solid state circuits but I have attempted to create a layout using the HRD's schematic of the circuit and would like to get some other eyes on it. Please let me know if you see any problems.

    Another issue, what's the best way to derive the +16 and -16 volts needed for the TL072 chip? Are there wall worts available? The HRD taps the power transformer's fixed bias secondary to create and filter the +/-16v power.

    I do plan to recommend a dip socket which isn't shown on the layout.

    [​IMG]

    [​IMG]

    [​IMG]

    [​IMG]
    A reverb pan and 8 pin dip socket are also needed.
     
    Last edited: Feb 25, 2016
  2. Tele-phone man

    Tele-phone man Tele-Afflicted

    Posts:
    1,951
    Joined:
    Sep 11, 2006
    Location:
    Asheville, NC
    Excellent work, but it begs the question: why are you even bothering to do this? The HRD reverb circuit is pretty dismal, unless you really like a fake sounding "BOING!" every time you hit a note. I've always thought it was one of their worst reverb circuits ever. Your layout looks great, but why would you want to build or replicate this circuit?
     
  3. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    Your attachments are just little blue boxes with a question mark with my browser. Yet your attachment in "Reverb Noise" comes through fine just like everybody else's. The attachments do work with Safari 7.1, but not with older ones.

    Also, your website elicits a certificate warning which keeps me, and possibly others, from viewing your website.
     
  4. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,260
    Joined:
    Dec 29, 2012
    Location:
    United States
    Ten Over, thanks for the info on my website, I'll look into it. Anyone else getting the certificate warning or not able to view the graphics? I don't have a problem with Chrome or Microsoft Edge browsers.
     
  5. mRtINY

    mRtINY Friend of Leo's

    Posts:
    2,173
    Joined:
    May 7, 2015
    Location:
    Orygun
    I thought Reverb drivers worked better as current-mode outputs, not voltage mode....


    -tINY
     
  6. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    Where are you planning to use this circuit? A lot of the time the signal going into the PI is too large for the non-inverting input of an opamp.

    I always thought those opamps made excellent constant-current reverb drivers, but I've never tried a HRD. They certainly work well in other production amps and in the ones that I have built.

    You don't necessarily need a +/- 16V supply. I don't know about the TL072, but some opamps can be tricked into thinking they have a split supply (and some can't). +30V is their positive rail, +15V is their ground, and amplifier ground is their -15V rail. Some of them, like the LM258, are intended to be used with a single supply.

    It is easy and cheap to make a split supply using a little 36V ct power transformer and at some point I decided this was the best thing to do.
     
  7. printer2

    printer2 Poster Extraordinaire

    Posts:
    7,090
    Joined:
    May 24, 2010
    Location:
    Canada
    So is it the circuit or the use of a short tank? Vaguely recall the amps coming with a short tank but later one's with a long one. Mind you, depends what you are looking for. I am planing on putting a tank in a Champ sized amp, a 17" length might be a bit much. Will this amp be my ultimate amp, hardly. But for messing around a little it might be fine. Have other amps in mind to build that might get a little more involved. Speaking of, thinking of putting a short delay on the signal before it gets to the reverb amp.

    (Almost forgot) Seems that there are aftermarket tanks that people like better than some of the one's Fender has used.

    The whole idea is to get the right frequency response, how you get it does not really matter if you get good results. Current mode or as Fender has been doing it for years, with a 6 dB high pass filter.

    For split supplies, lot of good examples on how to do it out there. A virtual center tap is one way. Whatever is the most convenient. I might even stack a pair of small switching 12v wallwarts.

    Another idea for the opamp, a pair of output transistors in the feedback loop if the need is there. Also remember, this circuit is for the 800 ohm tanks, the 2250 ohm may need more voltage, the 8 ohm tanks might be better served with a chip (car) amp running on 12V. Probably would be my choice if I didn't pick my tanks up cheap.


    4:12

     
    Last edited: Feb 24, 2016
  8. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,260
    Joined:
    Dec 29, 2012
    Location:
    United States
  9. mRtINY

    mRtINY Friend of Leo's

    Posts:
    2,173
    Joined:
    May 7, 2015
    Location:
    Orygun
    Interesting. Do you mean an RC network to give a 6dB/octave rising response?

    That would explain the lack of "drip" with the resonant frequency of the tank driver not getting driven as hard....

    That must be what R34/C14 is about in the schematic above.


    -tINY
     
  10. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,260
    Joined:
    Dec 29, 2012
    Location:
    United States
    I added the HRD power supply to the original layout in the first post.

    Here's the +16v / -16v power supply board as a standalone board to power your solid state devices from your tube amp's power transformer bias tap:

    [​IMG]

    HRD +/-16v DC Power Supply
    [​IMG]

    [​IMG]
     
    Last edited: Feb 25, 2016
  11. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    I don't recommend using the bias winding because most of them aren't designed to put out the current that this circuit is going to draw. Some designs use the B+ for the positive rail and the bias for the negative rail, but I don't recommend that either.

    Half wave? I don't like it. If you go with it, consider switching the positions of the 1000uf and the 100uf caps.

    I was going to attach a schematic, but it looks like I will have to resize it.
     
  12. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    I looked at the HRD schematic. It has a power supply that was designed for the IC's and it also provides the bias voltage as a secondary function. The typical bias winding is designed for bias only.

    Let's see if the attachment will fly this time.
     

    Attached Files:

  13. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,260
    Joined:
    Dec 29, 2012
    Location:
    United States
    Where are you getting half-wave rectification?
     
  14. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    It's a voltage doubler with full-wave from -16V to +16V, but half-wave from ground to either -16V or +16V.
     
  15. printer2

    printer2 Poster Extraordinaire

    Posts:
    7,090
    Joined:
    May 24, 2010
    Location:
    Canada
    See,
    http://www.talkingelectronics.com/projects/BasicElectronics-1A/BasicElectronics-1A_Page3.html


    I prefer full wave in principal but half wave works fine, just bump up the capacitance. Since we are supplying current for a couple of opamps, not like we are drawing a lot of current.

    As far as a winding being for the bias, if it is a dedicated winding rather than a tap it does not need to supply much current so the transformer manufacturer would probably go as small in diameter as he could for the wire. But there is a limit to how small he would want to go in terms of handling. A pain to work with real thin wire. So the bias winding might actually be able to supply more current than needed, might have a low enough IR loss to feed a couple of op amps.
     
  16. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    How much current do you think we are talking about here? The typical opamp power supply dumps most of the current until the opamp is really stressed. That way the opamp has a constant voltage supply while drawing a wide range of current.

    A dedicated bias winding with a typical bias circuit sags when you adjust the bias. So it is not only that you change the voltage divider, you also change the output from the winding. The change in current in this situation might be something like going from 1ma to 1.5ma. Imagine drawing an additional 68ma.
     
  17. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    Fender used a 500pf/1M RC network in a lot of their classic circuits. This is a first order high pass filter with the characteristic 6db per octave rise at the low frequency end of the bode plot up until it reaches the knee.
    I don't know what you are talking about. There is a resonant frequency with the input coil and the capacitor, but it doesn't have any significance here.
    R34 is half of the NFB loop. C14 has a value high enough that its impedance in series with the other components is not a factor in the lowest frequency of interest.
     
  18. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,260
    Joined:
    Dec 29, 2012
    Location:
    United States
    Let me correct your post, "but with the equivalent voltage output of a half-wave rectifier--from ground to either -16V or +16V."

    A single diode on a bias tap is an example of half-wave rectification because only half of the +/- AC voltage cycle is being turned into DC voltage. The SS power supply I use above has a second diode which allows the circuit to rectify both halves of the AC voltage cycle so it is a full wave rectifier. Now the voltage output is similar to a half-wave rectifier because of the voltage divider, but it is incorrect to call that half-wave rectification.

    It would be counterproductive to use a bridge rectifier in the power supply circuit above because that would give us twice the voltage output but with half the current. We don't need higher voltage, we only need 16 volts with max current available to drive the op amps. Why start with a higher voltage and lower current when we will have to use voltage dropping resistors to burn the extra voltage?
     
  19. Ten Over

    Ten Over Tele-Holic

    Posts:
    717
    Joined:
    May 13, 2015
    Location:
    Central California
    My post is irrefutably correct and, as such, needs no correction. We are talking about a voltage doubler and everybody knows that voltage doublers are half-wave at their midpoint.
    Not with respect to ground.
    Nonsense. Get out your little red pencil and trace the current.
    If you are talking about your power supply, you cannot use a bridge rectifier and still have the bias circuit. If you are talking about my power supply, it is designed to use the full secondary for the voltage and the current that I desire. Nothing counterproductive there.
     
  20. mRtINY

    mRtINY Friend of Leo's

    Posts:
    2,173
    Joined:
    May 7, 2015
    Location:
    Orygun
    Well - in the U2A circuit, the 470pF and 1M on the non-inverting input have a turn-over frequency of about 1/3 Hz. This seems to be a DC blocking filter...

    The network attached to the Inverting side of U2A causes the gain to be 1 at DC and fixes the gain to 101 at "high" frequencies. It's turn-over is at 154Hz.

    So, the linear response of this circuit looks like a simple HPF @ 154Hz as far as guitar signals are concerned. And it's a voltage-drive output unless the designer is depending on the TL072's marginal ability to drive 600 ohms....

    [​IMG]

    I did manage to find an impedance plot for an 8-ohm tank driver. There is no resonant peak at low frequencies (not enough restorative force from the springs, I guess).

    http://sound.westhost.com/articles/reverb.htm

    The difference between current and voltage drive, in light of this, is that the current-drive maintains the same current at higher frequencies. The voltage drive circuit will drive less current as the frequency increases. Current moves the springs in the tank.

    http://www.accutronicsreverb.com/main/?skin=sub05_05.html

    Sure, you can compensate for this in the recovery stage. But that will only even out the frequency response. The transient response of the driver will be different. That's why, I believe, you lose the "drip".


    -tINY
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.