Booster

[Rescyou]

I have a dc voltage of 0-1vdc and need to boost it to 0-5vdc with a minimum part count. Any ideas?

Resc.

 

[dingo]

I have a dc voltage of 0-1vdc and need to boost it to 0-5vdc with a minimum part count. Any ideas?

Resc.

what are you tring to do?

if it is a power supply it will be tricky, esp if the voltage is under 1v.

if it is a input to a circuit then it is very ez

 

[Rescyou]

I'm using a Pic controller to generate 12 channels of software based PWM output. By varying the cycles I can achieve the 12 channels flicker free signal for light dimming outputs without the use of an RC/Opamp filter. Unfortunately with the reduction of the cycles the output is just over 1vdc at maximum duty for each channel. Most drivers that I'm familiar with are the TTL ones thus the need for the boost to 5vdc. Basically in the end I need about 0-12vdc for the light dimming which is no problem to achieve if I can get 0-5vdc from the pwm outputs.

I can reduce the pwm channels and increase cycles to gain more voltage, but I was hoping that could find a simple hardware solution. I can get the 5v if I run the signal through a source driver then through a sink driver and then through another source driver and finally a mosfet, but that is an insane amount of parts and grief. I can also get the 5vdc fairly easy using other methods but unfortunately its 3-5vdc instead of 0-5vdc

I'm just being greedy with the PWM channels I guess and will most likely have to cut them down

Resc.

 

[Nigel Goodwin]

I'm using a Pic controller to generate 12 channels of software based PWM output. By varying the cycles I can achieve the 12 channels flicker free signal for light dimming outputs without the use of an RC/Opamp filter. Unfortunately with the reduction of the cycles the output is just over 1vdc at maximum duty for each channel. Most drivers that I'm familiar with are the TTL ones thus the need for the boost to 5vdc. Basically in the end I need about 0-12vdc for the light dimming which is no problem to achieve if I can get 0-5vdc from the pwm outputs.

I can reduce the pwm channels and increase cycles to gain more voltage, but I was hoping that could find a simple hardware solution. I can get the 5v if I run the signal through a source driver then through a sink driver and then through another source driver and finally a mosfet, but that is an insane amount of parts and grief. I can also get the 5vdc fairly easy using other methods but unfortunately its 3-5vdc instead of 0-5vdc

I'm just being greedy with the PWM channels I guess and will most likely have to cut them down

Resc.

I don't quite understand your problem, the PWM output from the PIC will be going from 0V to 5V - where are you losing it?. Perhaps you ought to post a picture of one channel of your PWM output.

A common way of getting the voltage you want (in your case 0-12V) is to use the PIC output to switch an NPN transistor on and off, the collector of the transistor is connected to 12V via a resisitor - giving a 12V square wave signal - filtering this will give a 0-12V DC output. This scheme is used in millions of TV sets and VCR's for the tuning voltage, but using a 33V supply (for 0-33V tuning).

 

[Rescyou]

With software based PWM each channel does not simultaneously output. Channel 1 turns on for 1ms then shuts off then channel 2 turns on for 1ms and then turns off and then channel 3 turns...etc.etc.etc. the full cycle is basically about 12ms long with each channel getting 1ms of it, thus the 1v output on each line. I was amazed that it put out that much as it should be only putting out about .42v (1/12 or 5v) I was hoping to amplify this signal to get the full TTL range.

I've used the transistor method you have stated but can't get pass about 2volts or the range is 2-4 volts etc. I've used 2803 darlington drivers with the same result. I've ran the signal through a pnp darlington then through a npn darlington and so far that has been the best boost then into a mosfet to get it up where I want it, but that is an insane amount of hardware to boost the thing and was wondering if there was a simpler way..

I was hoping to stay away from higher voltages then 12v but I'm going to try 24v on the transistors/darlingtons/mosfets.

Resc.

 

[Nigel Goodwin]

With software based PWM each channel does not simultaneously output. Channel 1 turns on for 1ms then shuts off then channel 2 turns on for 1ms and then turns off and then channel 3 turns...etc.etc.etc. the full cycle is basically about 12ms long with each channel getting 1ms of it, thus the 1v output on each line. I was amazed that it put out that much as it should be only putting out about .42v (1/12 or 5v) I was hoping to amplify this signal to get the full TTL range.

Right, so that's your problem - not so much doing PWM as outputting servo pulses. The first thing that comes to mind is doing proper PWM - you don't need to do them sequentially like you are - you can use interrupt routines to continually update the PWM outputs. If you consult the PICLIST at http://www.piclist.com there's code for doing multiple output software PWM.

 

[Rescyou]

Interrupts

I had considered utilizing interrupts to overcome the downfalls of my sequential way of pwm generation but I haven't learned much asm at the moment and it's going to take me a bit with MbasicPro to figure it out. I had previously read the stuff you are talking about but at the moment it looks more like klingonese than understandable code.. I'm pretty good at decyphering just about any language but asm is a large pain in the butt. It's been many yrs since I've programmed asm on an old Vax system that I'm sure is a boat anchor somewhere by now.. Oh the good'ole days of punchcards.....

Resc.