UHF Transmitter & Receiver

Thread Starter

Ryan M

Joined Apr 2, 2019
10
Hello, I'm looking for help on building a transmitter & receiver to replace the wired connections on my Electric Drum Kit. The drum kit currently passes a signal from the drum pad to the computer brain module when the pad gets struck by a stick via basic stereo or mono cables, the same cable you'd use to connect an mp3 player to aux-in of a stereo, either 3.5mm(1/8") or 1/4" jack cable like a guitar would use.

I'd just like to make these connections wireless as there are at least 15 drums and cymbals connected to the rack and each needs its own mono channel to send the signal to the brain module when its hit. The computer does most the work, The pad basically has a sensor called a "paizo" and I imagine it produces a small electric signal when the pad is hit at a certain threshold and sends it through the cable.

They currently sell 1/4" transmitter/receiver systems for guitars starting at around $40, which I'd be happy to use as a solution if one piece was enough, but like i said earlier each drum/cymbal has a dedicated channel so it would cost over $500 to complete this project (more than the kit itself costs). I've seen professional bands use these for guitars when on stage in a show. I also don't need a high quality reproduction of the signal like a guitar performer would need because this wouldn't be used for generating sound, it just needs to carry a a weak inaudible signal to fire the trigger. There are youtube tutorials on building these but the parts are still relatively expensive for they are building the ones with high sound quality reproduction.

I believe such devices for guitars are built using UHF and ive seen some at 900mhz probably because latency is the biggest issue when sending wireless signals, and in order to fool the human ear and be able to play along, latency needs to be below 11 ms i believe. UHF accomplished this i imagine.

Any help for the parts and schematics would be appreciated.

Thanks in advance!

-Ryan
 

Thread Starter

Ryan M

Joined Apr 2, 2019
10
I'm considering trying something like he attachment here, This is for a camera. It is to trigger the flash wirelessly. Depending on how difficult & expensive this project is, and whether something like this will work, I'm considering just purchasing ten of these to clean up the clutter unless someone has some ideas for how i can achieve what im looking to build. I dont mind if the project takes a long time because I'll take it on one cable replacement at a time. Perhaps 2 per month and that way it gets done no rush in the year 2019. I enjoy the task as well, and i treat it as a hobby :)
 

Attachments

Sensacell

Joined Jun 19, 2012
3,432
This is going to be more difficult than you imagine.

It's a bit like saying "I want to build a phone charger that can charge my phone 20 feet away"
The idea sounds attractive, it's technically possible under very limited circumstances, but not very practical.

Clearly, if this was technically easy to do, every E-drum manufacturer would have this feature, the benefits are obvious.
The trouble is that you must be able to handle simultaneous triggers, meaning that you need 10 non-interfering channels that operate totally asynchronously, with very low latency.

Really cannot think of any practical solutions.
The plug-in wires start to seem like a really sexy solution.
 

Ya’akov

Joined Jan 27, 2019
9,071
The signals from your drum kit are not going to be able to be directly transmitted via radio. For each piece of kit, you will need the front end processing the brain module is doing on the signal to be done locally, then the signal will have to be transmitted to a box outside the brain module that converts the signal back to what looks like raw drum kit output.

In addition, this will introduce, possibly not small amounts of, latency. The delay could really screw things up.

While this is theoretically possible, it is not practical. Maybe you could find a way to put the brain closer, and use wireless for the audio.
 

DickCappels

Joined Aug 21, 2008
10,152
Given enough thought this might turn out to be easier than it first appears. It would probably require some custom hardware and some programming.

How much latency can you accept? (The time between the drumpad being struck and receiving a signal from the receiver.)
 

Thread Starter

Ryan M

Joined Apr 2, 2019
10
Thanks for the comments guys! I agree that if it could be done then everyone would be doing it (That's my plan eventually, so who has a degree in marketing and who knows some investors LOL)

@DickCappels: Latency will probably be the deal breaker im guessing if anything, but research has show that 10 milliseconds or less is required to be able to play along accurately or else the audio will not feel linked to the pads being hit and the timing will be thrown off with any delays higher than 11 millisecs.

That said, I am working toward my degree in computer science so I have a fair amount of coding C++, Python & Java beneath my belt. So, If you have a solution with any amount of latency, I'm game to try it on one drum as an experiment, and will return with results and some pictures/video of the process and video w/ audio of the results showing latency and quality. If it works in the end I'd be forever grateful as I'll potentially be saving myself hundreds of hours over the coming years (I'll explain in a later post why mobility is a must and unfortunately its not just something that gets setup and left alone like I had originally anticipated with electric drums).

Thanks again guys in advance!kit.jpg
 

Thread Starter

Ryan M

Joined Apr 2, 2019
10
This is going to be more difficult than you imagine.

It's a bit like saying "I want to build a phone charger that can charge my phone 20 feet away"
The idea sounds attractive, it's technically possible under very limited circumstances, but not very practical.

Clearly, if this was technically easy to do, every E-drum manufacturer would have this feature, the benefits are obvious.
The trouble is that you must be able to handle simultaneous triggers, meaning that you need 10 non-interfering channels that operate totally asynchronously, with very low latency.

Really cannot think of any practical solutions.
The plug-in wires start to seem like a really sexy solution.
##
Thanks, and it's funny you mention the 20-ft charger because I have built one :) If all goes as planned Manufacturers will get onboard with my contraption in due time ;)
#I'm very optimistic :D
 

surri

Joined Mar 11, 2019
11
Here's an idea.

Wire short cables from drums to a box containing a microcontroller with 16 digital inputs and a nrf240x tranciever.
The microcontroller constantly polls the wires (<5 us per wire) to see if there's a signal.
When a signal is detected a 4 bit message (max 16 drums) is passed to the nrf240x and sent in shockburst mode.
(https://www.semiconductorstore.com/pdf/NewSite/nordic/WP_nRF240x_ShockBurst.pdf)

Another box near your "brain" with identical hardware will receive the 4 bits and straight away set the outputs according to the packet.

This will def have lower than 10 ms latency. Perhaps around 1-5 ms. But it will take some time and thought to design.
 

Ya’akov

Joined Jan 27, 2019
9,071
Here's an idea.

Wire short cables from drums to a box containing a microcontroller with 16 digital inputs and a nrf240x tranciever.
The microcontroller constantly polls the wires (<5 us per wire) to see if there's a signal.
When a signal is detected a 4 bit message (max 16 drums) is passed to the nrf240x and sent in shockburst mode.
(https://www.semiconductorstore.com/pdf/NewSite/nordic/WP_nRF240x_ShockBurst.pdf)

Another box near your "brain" with identical hardware will receive the 4 bits and straight away set the outputs according to the packet.

This will def have lower than 10 ms latency. Perhaps around 1-5 ms. But it will take some time and thought to design.
The signals from the drums are analog, the will require addresses and quite a few bits of amplitude. If you used TDM, it would probably run into the latency wall.
 

djsfantasi

Joined Apr 11, 2010
9,156
The signals from the drums are analog, the will require addresses and quite a few bits of amplitude. If you used TDM, it would probably run into the latency wall.
But what is “quite a few bits”? It’s not like s guitar. It’s a percussion instrument. This is a wild-assed guess, but do you need 255 levels to represent the amplitude? How about 64, 32, 16, or 8?

If your latency estimate is 1-5ms for a four bit address, is it reasonable to extrapolate that eight bits would take 2-10ms? Still within the TS specifications.

So, if sixteen levels of amplitude/ force/ loudness is acceptable, this is still a viable solution. Note that four bits is a “sweet spot”.

So? Are 16 levels of loudness sufficient?
 

Ya’akov

Joined Jan 27, 2019
9,071
But what is “quite a few bits”? It’s not like s guitar. It’s a percussion instrument. This is a wild-assed guess, but do you need 255 levels to represent the amplitude? How about 64, 32, 16, or 8?

If your latency estimate is 1-5ms for a four bit address, is it reasonable to extrapolate that eight bits would take 2-10ms? Still within the TS specifications.

So, if sixteen levels of amplitude/ force/ loudness is acceptable, this is still a viable solution. Note that four bits is a “sweet spot”.

So? Are 16 levels of loudness sufficient?
Most do MIDI range, 0-127 values.
 

Ya’akov

Joined Jan 27, 2019
9,071
Let me amend that, I say "most" but I was only involved with electronic drum quite a while ago. I am sure they haven't gotten worse, but they may well have gotten better.
 

Thread Starter

Ryan M

Joined Apr 2, 2019
10
Guys, This is INCREDIBLE feedback! I'm very thankful for the effort ya'll have put into assisting me. (11ms is the teetering point and we'll treat this as the ceiling).

I will do some digging and discover everything I can regarding the signal coming out. Please stand-by anyone who's not burnt out on this yet, and I'll return.

Thanks again! I'm super-excited for this experience !!

til then...
 

Sensacell

Joined Jun 19, 2012
3,432
I think the TS needs each drum /cymbal unit to be it's own separate, independent radio TX unit.

Am I correct?

This makes it harder, you cannot just collect all the signals at one point, process and transmit.
That's what the drum brain does. (I used to have a Roland E-drum set)
The TDMA idea has to be implemented at the radio level.
 

Thread Starter

Ryan M

Joined Apr 2, 2019
10
I think the TS needs each drum /cymbal unit to be it's own separate, independent radio TX unit.

Am I correct?

This makes it harder, you cannot just collect all the signals at one point, process and transmit.
That's what the drum brain does. (I used to have a Roland E-drum set)
The TDMA idea has to be implemented at the radio level.
Yes, this is correct. I was initially looking to produce 16 seperate RF transmitters, each with its own receiver plugged into an individual channel at the module. I'm imagining the same technology as a singer who uses a wireless microphone on stage, I just don't need the sound clarity or quality like a singer would, as the devices would be exchanging just small electric impulses. It just needs to pass whatever the signal is that normally travels from one end of the cable to the other end. Praying that I can somehow achieve the greatest cord-cutting project of all time and I realize if it was easy Roland would probably already be selling it in their top kits!

(by the way I had a Roland kit as well, before i got this Alesis and NOTHING compares to those Rolands, amirite? they kick a$$ and it costed like 5 grand; still paying it off lol)

Optimum goal would be to get all 16 channels wireless, but if i can only get 8 to work, then that would cut down half the wire clutter and worth something.

Thanks for the time and effort
 
Top