Reeded Plate Reverb. Direct Testing Implementing and Improving

Over the past day ive been testing out  my home made Reeded Plate Reverb unit directly on intrsuments whilst also recording a dry signal at the same time for mixing and comparison purposes.

The instruments i tested the unit on were, 

Amplified electric guitar,

Amplified Bass Guitar

Re - Amplified Vocals

Snare Drum 

The results from this unit on all the above instruments are very pleasing, especially on chords played by the bass and electric guitar, this leads me to believe that this unit would also sound great on a semi acoustic guitar on an amplifier. However the results were pleasing on the snare drum as it is loud enough to easily over come the noise issue this unit still has. The vocals sound ok, but they need to be louder to get the desired effect from the plat, but still dont sound to bad.

When testing the snare drum through the Reeded Plate Reverb i held the plate in different microphone positions, and i discovered the facing the plate to the wooden body of the snare drum gave the most desirable sound, but as with all microphone positions this is subject to opinion.

This certainly is the implementation for this unit. It does work as an effects send, but works so much better if its used at the recording stage as a microphone, and use in conjunction with a dry recording

To improve this plate and its effects i add another contact microphone to it o create a stereo effects unit, this improved the effect dramatically, adding more texture, and a greater sense of space, also making this unit much more useful in a mixing situation. However at first, adding another contact microphone did pose a problem.

The problem i en-counted, was that as the to contact microphones were in contact with the metal plate, it was creating a short circuit, therefore no signal was being received from the plate, however this was overcome by running each contact microphone through a separate DI Box, the separate transformers with in the DI box's help to create two separate circuits, thus solving the problem.

To improve the functionality of this unit i made a few physical adaptations.

These include,

A second contact microphone for stereo effects,

Feet, to make the unit self standing,

Clips, to make the unit attachable to microphone stands, this enables the user to place the plate in any number of microphone positions when recording.

In summary.

This is a 6 inch by 6 inch tin plate (shimming metal) with 24 reeds attached in a grid formation to one side. These reeds measure approximately 7.5 centimeters in length and 0.5 centimeters in width. The plate is attached to its aluminium frame via four springs. The tension on the plate can be adjusted by loosening two bolts and pulling or slackening the top of the frame. Two contact microphones are separately attached to the reeded side of the plate in two different positions, each as its own output to be placed in to separate DI box's or into a stereo Di Box. Feet and clips have been attached to improve functionality by allowing the user to place the plate on the floor or on a microphone stand close to the impulse they wish to use to excite the plate.

This is a working product.

Improvements?

There is a noise issue, this is due to the metal frame and lack of earthing. a wooden frame would improve on this problem but not solve it. A slightly larger plate with more reeds would increase the effect, however making the plate to large would make the unit cumbersome to work with, and less user friendly.

All in all, this plate has satisfied the intended brief, its a plate reverb thats had the successful adaptation of "reeds" added to it which positively effects the sound of the plate, by extended the reverb time and taming its resonance issue. It certainly adds a sense of space and depth to instruments it is added to.

It also proves my original theory behind this design was right, adding reeds do add to the reverb effect, as they add there own pitch and frequency resonances and characteristics plus by adding to the overall surface area they extend the reverb time of the plate.

My only hope, is that i get it back when my lectures are done with it. 

Its a very useful and effective unit. Well worth the build, time and effort. Its taught me a lot about plate reverbs, resonance and soldering. 

Reeded Plate Reverb.

After testing to prototype plate reverbs, one small sized standard plate just to test that i could send and receive signal to and from the plate, and the second just to test if adding tabs or "reed" like structures to the back had an effect, im now going to create what i set out to build, a Reeded Plate Reverb.

In the making of this reeded plate reverb, i cut 24 strips of tin (the same material as the plate) to approximately 0.5 cm in width x 7.5 cm in length. I then flattened each strip with a small hammer and bent over a small tab on a single end of the reed. This is the surface which will be soldered to the plate.

A third plate of tin was cut to exactly the same dimensions as the other two test plates (6 inch x 6 inch) and was marked out in a grid.

Each of the 24 reeds were soldered an inch apart in rows and columns of 5 reeds.

After placing the contact microphone on the plate (slightly off centre) and the plate back in the original frame with the original springs. Testing could begin with the same samples.

Result... SUCCESS! The plate actually sounds like reverb, all be it short, but there is a reverb effect that is audible. The reeds seem to have over come the pestering 400hz resonant frequency too.

Interestingly this plate seems to respond to higher frequencies better than lower frequencies. In my testing im using a clap, snare and bass drum sample, and this plate responds best to the clap sample were as the other two plates (due to there resonance problem) responded best to the kick drum or snare drum.

This is all because of adding the reeds, my theory behind this is that the reeds, soldered to the plate, not only add increased surface area but also interrupt the paths of standing waves across the plate, also the reeds create tension at the point that they are soldered to the plate creating different frequency characteristics.

Furthermore as the reeds are hand cut, shaped and soldered, no two are exactly a like, because of these small deviations in each reed (in terms of length, width, and the surface soldered to the plate) each reed does have  an individual pitch, even though they seem relatively similar in dimensions. I believe that this has the effect of thickening up the texture of the effect as so many different frequencies are effected due to the differences in each reed.

The signal from this plate is of much greater clarity than the two prototype plates due to its fuller and richer frequency content, however due to the greater amount of metal fixed to the plate, unfortunately the noise issue has increased, but mixed correctly it is not a major issue, but still has given me another idea.

I believe this noise issue could be overcome by a slight change in use. Instead of using this plate as an auxiliary effect off of a mixing desk, why not place this plate directly on the instrument you wish to effect, like you would a microphone. Especially drums as the loud transient sounds would require less gain. This could be the main advantage of this plate being small. Could it be used in conjunction with different microphone techniques and positions to acquire different characteristics? Why not?

Therefore im going to test this slightly different implementation and record the signal from the plate with it placed in a microphone position on a snare drum. But as it responds well to higher freqencies, what would it sound like on a guitar cab amplifying acoustic guitar, or a bass cab? I will test these implementations too. I will also record a dry sound of the instrument to use as a comparison, and to mix with the effect.  

Also i intend to add another contact microphone to the plate so that it can be used as a stereo effect, as this may also decrease the noise issue as less gain may be required, but also it may give the effect more presence in a mix. 

The 2nd Plate. Experiments prior to reeds.

After making a working plate, i decided to develop its design.

Next i decided to solder to 'tabs' to the rear of the plate.

The Theory is to increase the surface area of the plate to try and achieve a larger reverb time as with the existing plate being 6 inch's by 6 inch's with a surface area of 24 inch's (6 inch x 6 inch = 12 inch x 2 surfaces = 24 inch's ) there isn't enough area for reverb to occur.

Also this is a test to see if the sound is transmitted to the tabs, which are 2 and a half inch square in diameter which will add an extra 25 inch to the total surface area (2.5 inch x 2.5 inch = 6.25 inch x 2 surfaces =12.5 x 2 tabs = 25 inch's). This will also prove weather or not these tabs will resonate at a particular frequency.

To ensure this is a fair test the same speakers, frame, drum samples, springs and tension have been used.

The results of exciting this plate were pleasing. Not in the sense of sonic quality or a good effect but, the results proved my theory. The tabs added an extra resonance to the plate, however as the tabs are large in dimensions, unfortunately they produce bass frequencies, which for a reverb effect is not desirable.

The two tabs did not seem to increase the size of the reverb time, there effect was to lower the resonant frequency of the plate. To ensure this was true I placed the contact microphone on to one of the tabs and the effect was an increased resonant frequency which proved my theory.

The issues with this design are that still large amounts of gain re required so the signal is noisy. The plate is still not earthed which adds to the noisy signal. And the tabs are to large and therefore increase the low frequency resonance.

Next i attempted to build my own speaker system to excite the plate. For this i used a piezco driver with a capacitor soldered into the positive feed and screwed into a terminal block into which a guitar lead is used to drive the speaker.  

I decided to solder a capacitor into the circuit as this stops low frequencies (generally under 1khz) from being produced by the driver, the reasoning behind this is, if no low frequencies reach the plate, i should rectify any low frequency resonance. 

This driver had a rubber tube attached to increase its directionality and therefore improving its performance.

This driver was mounted close to the plate to ensue the maximum sound pressure level was exciting the plate.

I have since tested this design and it is a failure, this because the driver simply cannot generate enough sound pressure to excite the plate enough to receive a signal from the contact microphone.

I have since removed the speaker.

In my next design I intend to place 25 "reeds" which will be placed in a grid 1 inch away from each other and will be 0.5 cm in width and approximately 7.5 cm in length. I will then test this design to see what effect these 'reeds' have. Thy will increase the surface area by 66.96 inch's (0.5 cm x 7.5 cm = 3.75 cm x 2 surfaces = 7.5 cm x 25 reeds = 187.5 cm. To convert to inch's 187.5 / 2.8 = 66.96 inch's) which means the surface area is nearly 2.8 times greater.  

The total surface area = 66.96 + 24 = 90.96 inch's  

I am also hoping that the small width of the reeds will decrease the low frequency resonance and help boost harmonics across the length of the reeds.

I intend to trim each reed down to different lenghts as smaller reeds will resonate at higher frequencies and help counter the low frequency resonance. 

However a side effect maybe that the reeds act in the same way as a spring reverb as the sound will be carried down the reed, and reflect back of its boundary.

Thats all for now.

Testing continues. 

Plate Reverb Prototype Test.

Two day's ago i had the 1st opportunity to test out my prototype plate. The results were remarkable, as in it works, however it is not entirely pleasing.

As you can see in the picture's i soldered a contact microphone onto a terminal and then onto a guitar lead

.

This lead was then placed into a di box and routed through a mixing desk into a hard disk recorder, the plate was then excited by being placed in front of a speaker with a track being played through it.

Here are two sample recordings, they start with the dry track, then are a full wet signal from the plate, then the wet and dry signals are mixed together.

When you here the tonality of the plate Reverb sound change, that is the effect of moving the positon of the plate in front of the speaker, i.e were on the plate the sound is being focused, the closer to the middle of the plate the higher the tonality.

In the second Plate sample you can hear a dominant frequency, this is because the tension across the plate was increased allowing certain frequencies to travel faster across the plate. This gave a better effect than the 1st test with less tension, although this dominant frequency (between 400 - 500 hz) is a problem, this is due to the size of the plate and is a similar effect to modes and standing waves with in a room.

My plan to get round this problem is to attach reeds to the back of the plate which will increase the surface area theoretically combating the effect of the dominant frequency.

Each of these reeds will be different lengths, therefore they will have different resonant frequencies which will hopefully colour the sound of the plate and help mask this dominant frequency, by introducing different standing waves and therefore there harmonic content.

I am also going to use a slightly larger plate, and attach two contact microphones for a stereo effect, i will also need a microphone to capture the effects of the reeds.

Another noticeable issue with these first tests is that the plate is very noisy, this is due to an earthing problem, this can be resolved by placing the plate and frame in a wooden box and earthing the box, however this is only one reason to the plates noise issue.

To capture any useful signal from the plate, gain structures need to be incredibly high which introduces noise into the signal, to help with this i am going to drive the plate with a very directional pieczo horn, this will make the plate more efficient by exciting the metal more efficiently plus this horn will also eliminate any low frequency resonance as most horns cannot reproduce frequencies below 1khz

On evaluation, this plate reverb does not entirely work as a plate 'reverb' but more as a stereo widening effect, it is not what i intended to do but could become a reasonable achievment.

However i am hoping with the suggested modifications i will be able to create a compact stereo reverb.

Plate Reverb, Prototype

So, ive knocked up a small unit, using bits of aluminum for the frame and some car body paneling for the actual plate, fixed with springs to the frame. Ive purchased some contact mics and im in the process of stripping down some old speakers to se what driver will be suitable for this prototype. However, i have spotted some alarming problems with this design so far.

This prototype only has a plate that is 9 inch by 9 inch in diameter, this may cause the problem of feedback between the driver and contact mic, also finding a driver thats small enough but to fit on the plate and accurately recreate the signal its transducing may be a problem

Other problems consist of rattles and noise's from the frame, to solve this more tension and rubber grommets will be tested.

As for the reeds that are to be fitted to the plate, il be posting diagrams soon, but on this first prototype will not be equipped with these reeds, as i can use it to demonstrate the effect, if any, the reeds have on the side of the plate reverb.

Joe.

Decision Day

After research into this project, and looking at variouse different possibilities, including building a synthesizer, I have decided to build a plate reverb unit. But with a difference, im aiming to attach reeds to the back of the plate to see what effect this will have on the sound of the unit. Looking ahead im expecting the reeds to act like an amplifyer, for the frequencey at which they resonnate. But we'll wate and see.

The next stage is more in depth reasech, ive done some reading and now understand the basic concepts of how a plate reverb works, and i have a general idea of what materials will be best.

After some more in depth research im planning on making a small portotype, but with out the reeds, just to test the effectivness of the unit itself 1st, plus also il beable to compare the effectivness of the reeds and what effect they have more clearly if a have a standard unit to test the results against.

Stay Tuned.

- Joe.