Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: PRR on March 15, 2019, 11:37:04 pm

Title: Reverb design
Post by: PRR on March 15, 2019, 11:37:04 pm
http://roymal.tripod.com/reverb.htm
http://roymal.tripod.com/index.html
http://roymal.tripod.com/ultimate.htm
http://roymal.tripod.com/accutron.htm
http://roymal.tripod.com/faq.htm

Good basic work-up of reverb tank drive and recovery. Uses op-amps, but some concepts can guide tube reverbs.

(archived copies:  {cuz Tripod shouldn't still be working}
https://web.archive.org/web/20180225130337/http://roymal.tripod.com/reverb.htm
https://web.archive.org/web/20161027194817/http://roymal.tripod.com/ultimate.htm
https://web.archive.org/web/20180626223307/http://roymal.tripod.com/accutron.htm
https://web.archive.org/web/20120627192548/http://roymal.tripod.com/faq.htm  )
Title: Re: Reverb design
Post by: PRR on March 17, 2019, 04:32:06 pm
Interpolating from data above:

Hammond reverb tanks
Max peak I; and max peak V at 1KHz
-------------------------
10r   242mA   2.4V
190r  56mA  11V
240r  50mA  12V
310r  44mA  14V
800r  27mA  22V
1925r 18mA  35V
(basically 0.6 apparent Watts at 1KHz nominal impedance)
Title: Re: Reverb design
Post by: Colas LeGrippa on March 17, 2019, 04:49:53 pm
Very good post PRR but I don't speak chinese :l2:
Title: Re: Reverb design
Post by: 2deaf on March 17, 2019, 06:40:32 pm
I printed out that first article years ago and promptly contributed it to my local paper recycling effort.  It got off to a bad start when he claimed that a magnetic field imparts vibration to the unit's springs.  The little magnets in the gap rotate, they do not vibrate.

It really got thick when he claimed that Accutronics gave a spec of 25 A-T.  It's 2.5, NOT 25.  Look it up.  If I had gotten a drive voltage of 135V, I would have back-tracked in an attempt to find my mistake(s).  But he plowed on and claimed that a guitar doesn't put out Jack over 1KHz so that he only needed 13.5V.  You guys think a guitar doesn't put out much above 1KHz?  I don't.  This is just a ridiculous attempt to make an erroneous conclusion fit into reality.

He may well have some good points elsewhere in the article, I don't know.  I threw the baby out with the bathwater.

 
Title: Re: Reverb design
Post by: shooter on March 17, 2019, 08:06:37 pm
Quote
he claimed that a magnetic field imparts vibration
I know you slap a big 'ol pulse into a coil sittin in a 3t field, it'll make your eyes wiggle  :icon_biggrin:
Title: Re: Reverb design
Post by: PRR on March 18, 2019, 12:45:10 am
> Jack over 1KHz..... baby out with the bathwater.

https://quantumprogress.files.wordpress.com/2011/06/screen-shot-2011-06-10-at-3-45-51-pm.png

https://www.seymourduncan.com/forum/showthread.php?178456-Guitar-frequency-range&s=8d49ce35efc35a8a9c429909d0a10ab7&p=2327665&viewfull=1#post2327665

Title: Re: Reverb design
Post by: 2deaf on March 18, 2019, 10:57:27 pm
The first spectrum looks to me like a single plucked low G.  It actually doesn't have Jack above 1KHz.  I don't rightly play very many tunes that only have low plucked G's in them, although I may well be headed in that direction.
Title: Re: Reverb design
Post by: 2deaf on March 18, 2019, 11:04:29 pm
Now the single notes with distortion has all kinds of material above 1KHz.  Maybe even Jack.  I blew it up and strategically drew some lines in my favor.  The signals dropped a little over a half from 1KHz to 4KHz, but the impedance of the input coil increased by a factor of four.
Title: Re: Reverb design
Post by: 2deaf on March 18, 2019, 11:18:57 pm
Attached is a spectrum somebody posted somewhere of an Em chord with the neck pickup.  As expected, there is a lot of material above 1KHz and I think it is a pretty typical example.  If somebody took everything above 1KHz away from you, you wouldn't be a very happy guitar player.
Title: Re: Reverb design
Post by: PRR on March 19, 2019, 12:46:07 am
OK, I like your spectra.

> took everything above 1KHz away

NOT "take everything >1K away", but along a 20dB/decade line. Plot that line on your spectra.

I read it as clipping first on 240Hz (turn-down a hair), and also near 500Hz and 1KHz. Even under the 20dB/dec line there is 4dB headroom at 2KHz, 5dB at 3.7KHz, and more after that.
Title: Re: Reverb design
Post by: jjasilli on March 19, 2019, 08:30:16 am
Desperately seeking reality check.  How does all this actually sound?  For guitar?


We already know that guitars have only a few fundamental notes above 1000 Hz. Yes, the strings produce harmonics (overtones) above that.  And ea gain stage in the amp adds harmonics (maybe trimmed by NFB).  But these harmonics are weak.  Do they have the power to drive transducer & springs and make them selves heard to the listener?  Do we even want them to?


Traditional Fender tube reverb is fine, at least for guitar.  Maybe not for piccolo? 
Title: Re: Reverb design
Post by: 2deaf on March 19, 2019, 02:46:18 pm
The transformer-coupled tube reverb driver has a problem with clipping at low frequencies and at high frequencies.  You can see why this is by drawing some load lines.  At low frequencies when the AC load impedance is small, the AC load line is so vertical that it prematurely hits the zero current line resulting in clipping on that side of the signal.  At high frequencies when the AC load impedance is high, the load line is so horizontal that the other side of the signal clips over by the 0V grid line.  So it's like the low frequencies clip because the driver runs out of current and the high frequencies clip because the driver runs out of voltage.

The complaints you hear all the time are that low frequencies make the reverb sound "muddy" and high frequencies make it sound "shrill".  Maybe it's the distortion from the driver that is causing the objectionable sounds at those frequencies rather than the mere presence of those frequencies.  In any event, many driver designs attack the problem by cutting the high and the low frequencies.  The classic Fender reverb driver rolls off highs all by itself, but CBS saw fit to roll off highs even more by placing a capacitor between the plates and cathodes.

Anyway, those harmonics can make themselves heard to the listener if given the chance and I, amongst others, don't want them to.

It's all subjective, but maybe a piccolo could benefit from a little reverb? 
Title: Re: Reverb design
Post by: shooter on March 19, 2019, 04:05:12 pm
Quote
maybe a piccolo could benefit
:l2: :l2: :l2:
unlike normal folk, I don't get a "1st thought", I get an image, and I seen he poor 1776 fife guy dragging an amp with extension cord  :think1: