Jump to content

Strib

Members
  • Posts

    5
  • Joined

  • Last visited

About Strib

Recent Profile Visitors

588 profile views

Strib's Achievements

Newbie

Newbie (1/14)

  • First Post
  • Week One Done
  • One Month Later
  • One Year In

Recent Badges

0

Reputation

  1. Hello ! Thanks pete Of course I'm ready to share. The opamp/preamp come from the bobblick pages, he has developped this shematic (It can be publied in DIY projects, I've contacted him for the autorisation, that he nicely gave me. I thank him again) Then, I took the typical application of the TBA820M included in the datasheet : For making these schematics and PCB : http://www.strib.fr/images/sustainer/PCB/schema.gif http://www.strib.fr/images/sustainer/PCB/implant.gif http://www.strib.fr/images/sustainer/PCB/typon.gif (these are optimized for fitting into electronic cavity. This is the same circtuit you can view in my last post. => 400DPI images, print them at 25% scale) There is no more magnetic with the little neodymium, just a balanced magnetization of the bar. I've put 2 little magnets for the E/A/D strings, and 3 for the G/B/E . The leds are supposed to act as clip as you said. The problem I have, is that I must decrease the gain for harmonic mode, because I have an EMI noise if I switch with the same level than fondamental mode... Must solve this with a larger pot I think. I'll try to developp a little bit more the system this week end, before thinking about finishing the installation If you need more informations, feel free to contact me Strib
  2. Hi there ! Tests passed successfuly I've built 2 preamp/opamp to do my tests : - first fetzer ruby, which was not loud enough, because of the lm386 N1 chip (only 0.7w). As I was almost sure about the driver, then, I managed to build another amp, based on a LM741 preamp and a TBA820M opamp. The difference between the 2 circuits is very important, with only one pot for the gain, here it is : This is a prototype, I've now reduced the size of the PCB to 5cmx3cm . That will allow me to install it directly into electronic cavity. I've built 2 differents drivers, the first one with a piece of planer blade, 2.5mm thick : The second, made from transformer scratched : I first ued a ceramic bar from an old single coil, but the sustain was not very accurate, especially on the G/E/B strings (because of the diameter of the chords), then I replaced the bar with small square neodymium magnets (6mmx4mm 2mm thick), and this solved the problem. My old rusty guitar is now ready to cry : Here are 2 divs movies during first tests : fondamental mode harmonique mode Sustain forever Cheers, Strib
  3. Hello pete, thanks a lot for your quick response. I know that i'm on the right start, I'll be back with more experiment. I use tu wind my own pickups as I make my own guitars (you can have a look here : www.strib.fr ) Thanks again and, see you as soon as possible Strib
  4. Ok pete, but what is disturbing me is that the original bridge signal is splitted in 2 parts and go to the F/R input AND the guitar controls (I consider this wiring in "paralel"). Have a look at my scheme (Based on a typical wiring diagram from SD) : In this case : is the signal of the bridge pickup divided ? You pointed it ! I used a little guitar amp to do the job, but i know that I have to make the F/R soon (I'm waiting for some components right now) for improving my tests; When I told you that I have larsen and EMI, I had only the bridge pickup and the center single coil on the guitar. Neck pickup has been replaced by the driver. BTW, I plan to wind a special pickup like yours, which will have a typical 0.063mm (AWG42) wiring for the passive pickup stage, and a special stage above with 0.2mm wire for the driver, but it'll be another story ! Best regards Strib
  5. Hi there, I'm glad to join, many years after, this sustainer topic, and - first of all - i would thank every contributor for all the usefull informations we can find in this thread. I'm, at this point, building my own (already build my first driver prototype) made from some material i've canibalized from an old transformer (0.2mm wire), an old planer blade (core driver) some epoxy PCB and a ceramic magnet found on the back of an old pickup. I'm planing to build the F/R amp these days, but I did some tests with an external amplifier wich allowed me to have some sustain (fondamental and harmonical) on some open chords : Great start... I think I have to reduce the core height (mine is 5mm) to the ideal size defined by the psw's tests, and fine tune the position of the driver for the moment, there is a little thing I can't understand (even if I've searched for it in the entire thread ...) : How are connected the pickup wires which are supposed to be connected to the onboard guitar circuit AND to the F/R amp input ? I've picked the wires directly from the pots connections, but the signal (ie the final sound) of the humbucker i used, seems to be altered when sustainer is on. I have a totaly different attack, and weird sound. Thanks a lot for your help and cheers from Corsica strib
×
×
  • Create New...