23/08/2007 by peter.

After much faffing around i ported the PWM/RNG software to a PIC10F222 since I had some lying around and they are a little more suitable for this project. This involved a bit of wire sculpture between 2 DIP8 sockets to make an adaptor for the Low Pin Count board but was doable.

The next thing was to have a go at controlling a push-pull solenoid. The idea is to create a more controllable vibration generation system so that things like triaxial accelerometers or gyros can be connected to 3 orthogonally arranged vibration sub assemblies to give different sensations depending on movement of the user (or the movement of someone, or something else *insert imagination here*). An eccentrically wieghted motor generates a conical vibration pattern which has all sorts of problems with inertia and also vibrates in 2 rotational axes. One solution would be to get a motor which has a spindle out of each end (similar to those used in model rail, HO, OO scale would be a good start but those used in N and Z scales are tiny enough to have al sorts of possibilities) and attach a weight to each end of the shaft such that the weights are coaxial and that that axis is parallel to the axis of rotation. This would give vibration in 2 linear axis. Far better is to have an inertial mass being oscillated along one axis only. Which is where a push-pull, or more accurately opposed pull-pull, solenoid comes in.

As you can see from the above pic, the board with PIC (10F222), the psu board and a driver board (consisting of 2 MOSFETs, 2 1N4004s and 2 1M res) are connected by a handful of wires together and to a pair of solenoids attached to a base and whose shuttles have been wired together. The code was so short that it didn’t even go off the bottom of the screen (I say to all you people who write programmes in the Mb and Gb range, PAH!!) It takes in the ADC value from the input connected to the pot and uses that to change the time between and duration of the pulses going to the 2 solenoids making the combined shuttles shoot back and forth at a contorted rate. These solenoids are 12V and are not very good with the 9V battery I have to hand so a small lead acid bat will need to be ordered (and possibly some smaller solenoids)

next :

Since I’m doing this in totally the wrong order (more to do with what piece of technology i get my hands on next) the next fun thing is GPS. Just need to write some code to take the data from the GPS unit and then interpret that in a cleaver way, but as you see the thing is rather impressively small (reception inside a person may pose a problem so this may have to be connected to a small RF link like ZigBee or bluetooth and put in a location with more skyview, like a hat).

Anyhoo… Onwards!!!!

Posted in intelligent vibrator | Print | 1 Comment »

15/08/2007 by peter.

So after much tinkering and coding I have a motor who’s speed is random and the amount of time it runs for at any given speed is also random.

The photo below show the finished article. The Green PCB is a Microchip Low Pin count prototyping board (came with the PICkit2) which contains a PIC16F690. A header splitter is then attached to the port on the side of the board giving me the ability to plug 2 daughter boards into the main board. On this there is a motor drive board and a hardware RNG. The RNG has already been explained, the motor drive board has a diode (1n4148) coming off the PWM output pin going to a simple low pass filter made from a 1nF cap and a 100R resistor which then drives a Darlington voltage follower to run the motor. the Darlington pair is infact a bc547 and a TIP31A since the motor is drawing up to 300mA and 2 x bc547s had a tendency to get a bit hot. There is no fly back since the LP filter should stop di/dt from getting to high (in theory). The other board is simply a LM7805 giving 5V for the PIC and 9v for the RNG. The rest, as they say, is software.

Once my wee camera decides to work again, i shall take a video of the thing working.

Edit: 100_1584.MOV

Anyone wanting more info, leave a comment and I will do schematics and tidy up and post code.

Posted in intelligent vibrator | Print | No Comments »

12/08/2007 by peter.

Electromyograph electrodes. Since any company that sells them only sell them to research institutions so they look rather hard to get hold of as an private individual/mad inventor. BUT WAIT!! Thanks to the internet and those intrepid sexual adventurers we have e-stim, although I knew of its existence before, what i didn’t know/think of until I stumbled across it, they use the thing I’m looking for in reverse, these!! and what’s more they even sell the conductive gell. Well that’s that sourcing problem sorted out, just need to find a photoplethysmometer/Geer gauge.

Oh and the volunteer to try it all on.

Posted in intelligent vibrator | Print | No Comments »

12/08/2007 by peter.

After getting my little photo sensor (one of theses) attached to my PIC I started playing with it using a torch and my little finger i can sometimes pick up a pulse but it is very sensitive to movement and the pulse only shows up in the 2 LSBs. This problem may be alleviated somewhat when measuring form inside someone. Anyhoo that is a problem to be sorted when I have a willing subject to test on.

But that aside, the electromyograph idea still needed a bit of thinking before anything could be built and I had no Idea if the output from it could be useful. Then I found this. Looks like we may be onto something here. All I need now is the person and the electrodes. Oh and the amplifier hardware. Oh and a data logging device. Hmmm… This is turning into a bit of a big project.

Posted in intelligent vibrator | Print | No Comments »

11/08/2007 by peter.

Here you can see my wee set-up. the LED sits there and changes its brightness about once a second (software settable up to about 30 secs) the board with the 9V bat is the RNG using built from 3x 2n2222, 3 resistors and a .1uF cap.

Posted in intelligent vibrator | Print | No Comments »

11/08/2007 by peter.

So after a week away and doing far less coding than I anticipated, I have finished the first programme and prototyped the hardware (well most of it). I wrote the code in MPLAB and used MPSIM to simulate it then used the PICkit2 to do the real world testing. It takes a hardware RNG intput, a PIC16F690 (a bit of over kill but it came with the kit and gives plenty of room for improvements and add ons) uses the rendom number with accumulator, ignores the overflow and then shifts 2 places left to give a nice random looking string of numbers since my RNG hasn’t got the right amplification at the moment to give a full scale for the ADC. This is then used to control a PWM loop (until I can work out how to use the PWM port). A couple of other loops hold the output at a level for a set time and then it repeats. the next stage is to run a motor off this output with a single transistor follower, work out how to set up the on board PWM port and do a quick amplifier for the RNG to give a better output.

Since no one has commented on this blog there is probably no one reading it. Therefore if anyone is acctually reading it and interested in the code, let me know and I will post it.

Posted in intelligent vibrator | Print | No Comments »

04/08/2007 by peter.

Temperature measurement.

Posted in intelligent vibrator | Print | No Comments »

04/08/2007 by peter.

Well after much procrastonating and thinking of “Am I really going to prototype a sex-toy?” I decided to bite the bullet (pun intended) and start code bashing. Starting with the PWM code I wrote for an RGB colour changing light and modifying it. I wrote a very long winded bit of code to pull random 1s and 0s from a pin and convert that into an 8 bit number and bolted that to the top. Removed the GB bit and renamed variables. I was doing this at 1am while listening to Radio 2’s “Friday night is music night” which was a special birthday celebration for Barbara Windsor which I had bee at the recording of. Not sure which bit of that was the weirdest bit. listening to show tunes while writing code for a vibrator or listening to something on the internet that I had seen recorded. Anyhoo, The new PIC programmer that I ordered to implement this (along with other projects) however looks like it won’t arrive until next week when I’m not here. So full testing of code will be (hopefully) commencing the week after. Once all verities of code all work (or if someone wants to see it before it all works) I will post it for those of you with buzzing hardware to integrate it into as I don’t have any one to beta test this mad idea on or to take measurements form to build the photoplethysmometer or electormyomometer attachment.

Posted in intelligent vibrator | Print | No Comments »

02/08/2007 by peter.

I wandered slightly off topic for a bit but now back on the path to the ultimate Infur-O-Tron (from infuriating)

It seams that anything you care to devise has already been constructed by someone else for some other purpose. For example, this unit, correctly connected up could give hours of enjoyment to an engineer with all the possibilities of the data gathered and then the fun of generating feedback circuits, to then give hours of joy to the user, taking them to the brink and dangling them over by an ever so fine, yet perfectly proportioned, thread. Assuming the engineer has done their maths correctly correctly, the hours could become an indefinite thrill ride.

Something else to knock up in an odd moment and then try and find someone to test it.

Also in the vein of the last post, gyros and accelerometers are cheap and small now, how about a unit that only works if your standing on your head, or vibrates with acceleration, or if you turn round. or only works if you reach a threshold of acceleration.

Or for those of you on the look out for random systems, how about a little sensory fusion with all these sensors, one big algorithm to give an out put and then depending on where you are, the ambient light, you blood oxygen level, heart rate, radio signals in the area, temperature, galvanic skin condition, acceleration, ambient noise and rotation, you get a different response. so standing in one place you could get pulses of vibration with your heart beat, in another place it could be light levels turning it on and off and the proximity of a mobile phone that modulates it or it could go off when some one talks to you. Oh for the beautiful maths that could do all that. And the ability to write it to a uC. And the equipement to turn it into reality.

one can but dream….

Posted in intelligent vibrator | Print | No Comments »

02/08/2007 by peter.

In this world of abundant information, there is a bounty of possible sources of data to use in intelligent vibrators. I started to wonder if a GPS Receiver Module were attached to the uC controlling the vibe then it would only operate at specific pre-set locations (dom/sub fun again) or at regular intervals intervals (say every 5 yards) so that being a passenger in a car or some other mode of transport could be made more interesting.

Or if your looking for something a bit more interesting, take Wardriving to another level. Using a WiFi Spectrum Analyser you could have a vibe that goes of when theres WiFi about. Brings a whole new meaning to “HotSpot”.

no doubt I’ll think of something elses

Posted in intelligent vibrator | Print | No Comments »