Positional Sensing Board dramatically improves input quality

[Alek_A]

While adjusting the settings I found out that half of my pads have so limited dynamic range and I can easily get highest velocity with a slight (not hard) hit. HighLevel set correctly, if I set it higher I just get velocity lower than 127 no mater how hard I hit. Even my four Roland PDX-100 pads behave differently, first three - perfect, the rest is bad. Of cause I tried swapping them, swapping the cables, but nothing changed. It became absolutely clear that MegaDrum's inputs are physically different.

I was disappointed because the cheapest cymbal CY-5 has more or less good results on all inputs. But two crash CY-14C-MG cymbals that is 5 times more expensive than CY-5 was so limited...
Situation dramatically improved when I connected them to Tom inputs! The crash cymbals shown a doubling dynamic range. And not only...

While testing I found out that these brilliant inputs reside in range from 4/5 to 12/13. Took a quick look at the PS board. And it's clear. These inputs go through the chips on the board. Others connected directly.
I also found confirmation and deeper understanding of positional sensing in the forum. viewtopic.php?f=3&t=1215

I can confirm that pads connected to the PS inputs (4-12) behave MUCH-MUCH better! I just wonder why has nobody mentioned this???
Less noise!
Less false triggering!
Accurate adequate dynamic response! (I mean when hitting with half velocity and below non-PS inputs are well too, but if you hit harder - sadly to admit, they give a portion of random)
Extended dynamic range! (x2 for CY-14C-MG, ~x1.5 for PDX-100)
REAL positional sensing! Unlike just almost random numbers I get on non-PS inputs. (Just saying what I've got with my PDX-100, maybe somebody has better results)

In my humble opinion PS board simply shouldn't be an option. It's a must have. We can't compromise quality, having such a great solution!

Tested pads are CY-15R-MG, CY-14C-MG, CY-5, PDX-100. Maybe somebody wants to post what effect the PS-board has on other pads!! Please Post!

Dmitry, I wonder if you could possibly do me a huge favor and design new PS-board with all inputs at least up to 35 (first 40 pin connector) PS-enabled.
And I would be happy if it had the same form factor, because my megadrum module too tight and simply won't take in bigger one. Could you?
Or at least direct me!

Best Regards!
Alek

[airflamesred]

While adjusting the settings I found out that half of my pads have so limited dynamic range and I can easily get highest velocity with a slight (not hard) hit. Alek

Therefore your pads are too hot.

[dmitri]

I was disappointed because the cheapest cymbal CY-5 has more or less good results on all inputs. But two crash CY-14C-MG cymbals that is 5 times more expensive than CY-5 was so limited...
Situation dramatically improved when I connected them to Tom inputs! The crash cymbals shown a doubling dynamic range. And not only...

While testing I found out that these brilliant inputs reside in range from 4/5 to 12/13. Took a quick look at the PS board. And it's clear. These inputs go through the chips on the board. Others connected directly.
I also found confirmation and deeper understanding of positional sensing in the forum. viewtopic.php?f=3&t=1215

I can confirm that pads connected to the PS inputs (4-12) behave MUCH-MUCH better! I just wonder why has nobody mentioned this???

My tests with pads and cymbals I have (Yamaha PCY-135 cymbal, Pintech cymbal and 10" snare, Roland PD-125X, Drumtec 14" snare and some no name pads/cymbals) showed no noticeable difference in sensitivity, dynamic range or triggering quality between inputs with or without PS precision rectifiers. Of course settings are different when using the same pad/cymbal on inputs with or without PS precision rectifiers since the PS precision rectifier "cools" the connected pads/cymbal and may also modify signal's strength curve.
Can you post settings for CY-14 on a input without a PS rectifier and describe the problems in details?

Dmitry, I wonder if you could possibly do me a huge favor and design new PS-board with all inputs at least up to 35 (first 40 pin connector) PS-enabled.
And I would be happy if it had the same form factor, because my megadrum module too tight and simply won't take in bigger one. Could you?
Or at least direct me!

You can take the schematic from viewtopic.php?f=3&t=1215 and design a PS board as tight as possible for as many inputs as you need. For the form factor of the PS boards I build I cannot imaging squeezing more IC/parts to increase the number of inputs.

[Alek_A]

While adjusting the settings I found out that half of my pads have so limited dynamic range and I can easily get highest velocity with a slight (not hard) hit. Alek

Therefore your pads are too hot.

Thanks for your feedback, airflamesred!
As Dmitri staid:

I mentioned it many times on the forum that if with HiLvlAuto set to Yes and Gain set to 0 after a dozen or so very hard hits HighLevel goes above 900 then the pad is too hot. Anything between 300 and 900 is good.

Additionally setting HighLevel to 64 before test, after sequence of some really hard hits I got HighLevel 575 for PDX-100 and 571 for CY-14C-MG. So I suppose these pads are not too hot.
Even knowing that, I tried this scheme: https://www.megadrum.info/~megadrum/for ... 441#p13441, but as expected it didn't help. As I remember, dynamic range boundaries just shifted.

[Alek_A]

My tests with pads and cymbals I have (Yamaha PCY-135 cymbal, Pintech cymbal and 10" snare, Roland PD-125X, Drumtec 14" snare and some no name pads/cymbals) showed no noticeable difference in sensitivity, dynamic range or triggering quality between inputs with or without PS precision rectifiers. Of course settings are different when using the same pad/cymbal on inputs with or without PS precision rectifiers since the PS precision rectifier "cools" the connected pads/cymbal and may also modify signal's strength curve.
Can you post settings for CY-14 on a input without a PS rectifier and describe the problems in details?

Sure. I think attached diagram describes the problem in the best way..

PS-vs-non-PS-mapping.png

As I said in the lower half of the range there's no problem, but in the higher a bit of random added and most unpleasant - limitation. If the range was not limited then I would try to correct it with the appropriate curve. But since it is - no go.
Settings in attachment!

Alek-Settings.zip

(those ending with "-PSI" for PS inputs, others for non-PS ins)

You can take the schematic from viewtopic.php?f=3&t=1215 and design a PS board as tight as possible for as many inputs as you need. For the form factor of the PS boards I build I cannot imaging squeezing more IC/parts to increase the number of inputs.

Thank you! But unfortunately I'm so far from circuit design, I see myself spending days on understanding the basics and designing it.

[dmitri]

Can you please just post the screenshot of settings?
Also I struggle a bit to read the graph since it shows how "hardness" depends on "velocity" not the other way around as it should be. The way you presented "hardness" is limited but "velocity" can go up to infinity.

[Alek_A]

Absoloutley right! My mistake. Updated graph. Settings in attachment.

[dmitri]

Judging by the graph and settings for the non-PS input you need to raise HighLevel.
If after raising HighLevel to 1023 you still get maximum velocity with less than hardest hits then you need to drop Gain by one and start from HighLevel=64 again (you will also want to lower Threshold to keep sensitivity the same with lower Gain). And so on until Gain 0. If it's still the same with Gain 0 then you need to "cool" the cymbal with a voltage divider.

[Alek_A]

Actually, I have set HighLevel quite precisely. If I raise LighLevel even by one, I won't get 127 velocity no mater how hard I hit, I will get 126.

[dmitri]

Nevertheless, can you do the same again with Gain 3 and post the settings?