Strange reverb feedback problem ... even stranger solution

Make it loud here.
Post Reply
User avatar
sookwinder
Mods
Mods
Posts: 10854
Joined: Mon Sep 11, 2006 6:47 pm
Location: Melbourne Australia

Strange reverb feedback problem ... even stranger solution

Post by sookwinder » Fri Aug 31, 2012 10:48 am

I have been happily playing/recording with the final version of my IMPRO REVERB amp now for 3 months after designing/building it earlier this year. 32 watts of pure EL34 bliss, with Fender style tonestack/reverb/tremolo. All in the small footprint of a Princeton Reverb cabinet
My 14 year old nephew has also been "blowing away" others at his school with his version of the amp he built.

However I have noticed every now and then a strange feedback issue that occurs in the reverb circuit. In fact is is only in the reverb recovery part of the circuit, I can remove the 12AT7 from the first part of the reverb circuit and the problem still occurs.

The FB is somewhere in the range 600 - 800Hz and once it starts , usually after playing a single note in a run, the only way I have been able to stop it is turn off the amp or bump/move the tank. Not the perfect solution if in front of an audience. As my nephew had experienced similar problems (all be them less frequent than I had) I started to think about this from an problem solving perspective.

So after some experiments which included connecting a tank outside of the cabinet, connecting to another reverb tank located in another amp, building a "Faraday cage" to put the tank in, fitting a smaller 25 watt speaker into the cab, changing the orientation of the reverb tank, I came to the conclusion that the problem was magnetic feedback, not the normal electrical or acoustic/vibrational feedback.

Because the amp is a "pocket rocket" and uses a 50 watt speaker, the magnet is far closer to the reverb tank than what would normally occur in a standard Princeton Reverb set up. So the feedback starts and is magnetically inducted by the coils in the reverb tank. Move the tank 50mm back away from the speaker ... no issue. Move it closer to the speaker...even more feedback.

This shows how close the speaker magnet is to the reverb tank
Image


So what to do ? what to do indeed? I didn't really want to get another cabinet slightly taller as i really like the compact style of the PR size cabs. So I looked in a totally different direction.

Rather than try and fix the distance issue between the tank and the speaker, I decided to look into magnetic shielding
and sourced some Mu-Metal Foil, which is designed for shielding low-frequency electro magnetic fields.

MU Medtal is an alloy of nickel, iron, molybdenum (80% molybdenum)

Order some sheets and it arrived two nights back.
Image

My idea was to make a (origami style) box for the tank to sit inside, which in turn the box and tank would be put into the reverb tank bag as per normal
Image

I had visions of having to strengthen the foil with a backing of some sort , so the "box" would not get crushed inside the reverb back. However when the foil arrived it was obvious this would not be required . 0.2mm thick steel sheet is pretty stiff and hard to bend !!! So I cut out my origami box
Image

Then folded it, with the aid of a hammer and a hunk or steel I had handy.
Image

By this stage it was 10pm and I had not realised that I had not even had dinner yet... decided to just put the tank in the box, then everything into the tank bag and see if it solved the issue. Turned on the amp... played a few runs and .. and .. shit ... the feedback was still there.

Annoyed, I went off to make dinner and think about the issue.

Came back an hour later with the notion that I should cover up / fit the final end of the box, just for completion sake. So I made the final end piece, fitted it to the box and repeated everything. Turned on the amp, played away and YES.. no more magnetic feedback !!!! I had solved the problem. I sort of remembered by second year electronics classes that talked about when shielding magnetic field you have to ensure all 6 sides are covered otherwise you are wasting your time... well those old professors were correct.

While I had the back of the amp off, I decided to fit a Weber 10F150-O smooth cone speaker that I had had for a couple of months. I had been running it in to see if it would loosen up a bit. Well it has definately loosened up and sounds great... seems it needed the 20 or so hours of break in to start to sound good.
Image

put everything back and the amp is perfect again.
Image
relaxing alternative to doing actual work ...

User avatar
TalkingBook
PAT. # 2.972.923
PAT. # 2.972.923
Posts: 253
Joined: Thu Jan 05, 2012 8:16 am
Location: Osnabrück, Germany

Re: Strange reverb feedback problem ... even stranger solut

Post by TalkingBook » Fri Aug 31, 2012 12:15 pm

Great looking amp (BTW, have a build thread/some frontal shots?) and good thinking with the fix. Thanks for sharing!

How much was the mu-metal?
Stuff: CIJ JM, Laney VC30, Highway One J-Bass, Markbass Little Mark Tube & Club 102

Band: litus.bandcamp.com

User avatar
OffYourFace
Mods
Mods
Posts: 12400
Joined: Thu May 03, 2007 2:59 pm
Location: Tarzana, CA

Re: Strange reverb feedback problem ... even stranger solut

Post by OffYourFace » Fri Aug 31, 2012 6:10 pm

all the new tanks are microphonic... i think it's due to the thinner sheet metal. We use the Mojo tanks. out of a box of 50, i usually have to send back 25. It's completely sad that accutronics are no longer made in the USA. The black korean tanks sounded poor in the amps and were just as microphonic. The Mojos sound ok.

We shield the bottom of the cabinet and run a wire to the tank's chassis. I put the tanks back in their cardboard box and then cover them with the typical fender reverb bags.

User avatar
sookwinder
Mods
Mods
Posts: 10854
Joined: Mon Sep 11, 2006 6:47 pm
Location: Melbourne Australia

Re: Strange reverb feedback problem ... even stranger solut

Post by sookwinder » Fri Aug 31, 2012 8:21 pm

TalkingBook wrote:Great looking amp (BTW, have a build thread/some frontal shots?) and good thinking with the fix. Thanks for sharing!

How much was the mu-metal?
`

here's the link to the build thread:
http://www.offsetguitars.com/forums/vie ... 10&t=59259
here's a link to the place in the usa where I bought the MU METAL
http://www.lessemf.com/magnetic_shielding.html
the cost is $32 per linear foot (12" x 15") for 0.01"thick (for those of us who are metric 375mm x 300mm x 0.25mm)
one linear foot was enought to build the box for a small reverb tank... you'd need 2 linear feet for a normal/large tank.
The biggest cost as far as I was concerned was shipping... costing the same as the item I purchased !!

In Europe there is this guy selling a similar product (the product is from Germany)
http://www.ebay.co.uk/itm/Mu-Metal-Foil ... 1c2a504897
OffYourFace wrote:all the new tanks are microphonic... i think it's due to the thinner sheet metal. ..........
The black korean accutronics tanks sounded poor in the amps and were just as microphonic. The Mojos sound ok.
Mike ... I have done a lot of back to back tests with the various long/short/2 spring/3 spring Korean accutronics tanks and while I will concede that they probably are not as good as a vintage Fender tank (inparticular the one in my 1966 PR) my impression is that they are OK...even venturing to pretty good.... great wet splashy reverb or subtle reverb ... all can be acieved without grittiness .. .. FWIW in my amps they certainly sound better than my 2005 reissue fender reverb unit !!!! (which now just sits there gathering dust)

In the 5 amp builds I have done prior to this one I have had no issues with the accutronics (korean) product .. but then I also have a slightly different reverb recoverey circuit using a whole 12AX7 ... more than what most mass production manufacturers use...

However... and it is a big however ... I do not take my amps gigging, they stay in a studio/home, also as i have discovered and as you have pointed out, there seems to be a cost down in the production process in having thinner sheet metal in the tank itself.... which may have contributed partially to the issue of the feedback I was experiencing.

Interesting your comments about the DoD percentages of MOJO tanks as well
relaxing alternative to doing actual work ...

User avatar
Langwich
PAT. # 2.972.923
PAT. # 2.972.923
Posts: 317
Joined: Fri Oct 07, 2011 4:19 am
Location: SFV, SoCal

Re: Strange reverb feedback problem ... even stranger solut

Post by Langwich » Fri Aug 31, 2012 8:29 pm

sookwinder wrote:I have been happily playing/recording with my IMPRO REVERB amp now for 6 months after designing/building it earlier this year. 32 watts of pure EL34 bliss, with Fender style tonestack/reverb/tremolo. All in the small footprint of a Princeton Reverb cabinet
Is there a thread on this build?

And . . .
OffYourFace wrote:all the new tanks are microphonic... i think it's due to the thinner sheet metal. We use the Mojo tanks. out of a box of 50, i usually have to send back 25.
A box of 50?! Do you guys have a production line going on?

User avatar
sookwinder
Mods
Mods
Posts: 10854
Joined: Mon Sep 11, 2006 6:47 pm
Location: Melbourne Australia

Re: Strange reverb feedback problem ... even stranger solut

Post by sookwinder » Fri Aug 31, 2012 8:40 pm

here's the link to the build thread:
http://www.offsetguitars.com/forums/vie ... 10&t=59259


Mike works at Matchless (while his mansion in the Riviera is being painted :D )
relaxing alternative to doing actual work ...

User avatar
OffYourFace
Mods
Mods
Posts: 12400
Joined: Thu May 03, 2007 2:59 pm
Location: Tarzana, CA

Re: Strange reverb feedback problem ... even stranger solut

Post by OffYourFace » Wed Sep 05, 2012 10:44 pm

We bought about 10 of the MOD tanks from AES. They sound great! But they only work in our heads ATM. I tried 4-5 of them in a Lightning reverb and each fed back when the reverb knob was dimed. The weird thing was, each tank fed back the same note! (D3) Even when I hooked a few up to a DC-30 reverb. That's a new one ???

Anyway, I really liked the sound of them. I bet they sound amazing in Fenders. They had a way nicer bloom than the Mojo tanks, IMO.

User avatar
s_mcsleazy
PAT. # 2.972.923
PAT. # 2.972.923
Posts: 16080
Joined: Thu Jul 28, 2011 7:30 am
Location: glasgow

Re: Strange reverb feedback problem ... even stranger solut

Post by s_mcsleazy » Wed Sep 05, 2012 11:52 pm

the amp is sooooo blue
offset guitars resident bass player.
'Are you trying to seduce me Mrs Robinson? Or do you just want me to solder a couple of resistors into your Muff?'

User avatar
maximee
PAT. # 2.972.923
PAT. # 2.972.923
Posts: 799
Joined: Mon Mar 05, 2012 2:07 pm
Location: Germany

Re: Strange reverb feedback problem ... even stranger solut

Post by maximee » Thu Sep 06, 2012 12:25 am

Thanks for sharing this. Very interesting and nice fix! Stunning amp you built there!

User avatar
sookwinder
Mods
Mods
Posts: 10854
Joined: Mon Sep 11, 2006 6:47 pm
Location: Melbourne Australia

Re: Strange reverb feedback problem ... even stranger solut

Post by sookwinder » Thu Sep 06, 2012 1:17 am

OffYourFace wrote: The weird thing was, each tank fed back the same note! (D3) Even when I hooked a few up to a DC-30 reverb. That's a new one .
That was the same note that was feeding back on the tanks I had from Acutronics ... weird !!!
relaxing alternative to doing actual work ...

User avatar
maximee
PAT. # 2.972.923
PAT. # 2.972.923
Posts: 799
Joined: Mon Mar 05, 2012 2:07 pm
Location: Germany

Re: Strange reverb feedback problem ... even stranger solut

Post by maximee » Thu Sep 06, 2012 4:53 am

Haven't really thought about it in depth, but to my limited understanding of coils/magnets it makes sense that these tanks feedback/resonate at the same frequency if they have the same springs of equal length. I'm really interested how that feedback sounds... could be interesting to play with.

User avatar
Motrcycleboy
PAT PEND
PAT PEND
Posts: 1
Joined: Tue Aug 13, 2013 5:37 am

Re: Strange reverb feedback problem ... even stranger solut

Post by Motrcycleboy » Tue Aug 13, 2013 8:47 am

Hi

I am currently experiencing the same problem with my Deluxe Reverb Reissue amp. When the reverb is quite high, say 8 - 10, I get feedback which starts like a low whine but gets gradually louder. Sounds like someone running a finger across a wine glass.

I am thinking of trying the Mu metal solution. You say that it is important to cover all 6 sides. Sound like a silly question but did you just leave a small gap for the RCA leads to come out one side of your box? I take it a small gap shouldn't let any magnetic interference to occur.

There is a company that sells Mu metal near where I live so an going to try them. Not sure my origami skills are up to it though so am going to see if they sell it as a sheet so I can wrap it around the tank.

John

Post Reply