Sorry, I forgot to mention that I need the about 50% duty cycle along with the shorter wave. That would explain why I was confused with yours. Like in the first picture I posted, I need both the waves and the shorter one always directly following the larger one.

Anyway, I'm having a lot of trouble with this and haven't gotten any where.

 

Sorry, I forgot to mention that I need the about 50% duty cycle along with the shorter wave. That would explain why I was confused with yours. Like in the first picture I posted, I need both the waves and the shorter one always directly following the larger one.

Anyway, I'm having a lot of trouble with this and haven't gotten any where.

Since that's the case it looks like you're back to a solution external to the 555. So back to the NPN aproach....

You need a more robust selection of components when doing experiment and design. Having only 1000uF and 10uF for C1 values is not going to cut it. When I said "Try reducing the value of C1", I didn't mean that drastically. I was thinking more in the order of 500uF. If you have two 1000uF Caps, you can put them in series to get 500uF. A FET would make things much easier because you could work with very small values of C1, but hey, I'm repeating myself.

I realize that if you're a young fellow you may not have the finances to purchase whatever you want. If that's the case you would be well advised to learn the art of cannibalism. Dumpster diving and curb side scavenging can be an art form.
Note: Hospitals are a great source for discarded electronics. Just contact someone in the Bio-Med Electronics, Engineering, or IS departments. Many will be eager to help a young electronics enthusiast.

 

Here's a modified version of the NPN circuit. The first stage is an Emitter Follower Amplifier, also called a Common Collector Amplifier. Note that with this scheme C1 can be much smaller, so your 10uF Cap should work OK. The graph shows the output differences when changing R2 to the values indicated. This way you can choose what 'Out" time you want.

Edit: Added D2. It will prevent Q1 base from exceding -.7V when input goes low.

 

If the circuit needs to recover faster on the negative edge, or the input amplitude is greater than 6V p-p, you should add a diode, anode to GND, in parallel with R2. If the input can sink a lot of current (>100mA?), it might be prudent to add 1kΩ in series with the capacitor, to limit the transient current.
The reason for mentioning 6V is that the base-emitter junction of Q1 will break down with higher voltage unless the diode is in place to limit the negative swing. Admittedly, the breakdown current would be low, but I am a conservative circuit designer.
Also, the diode will slow down the fall time of the negative edge if the resistor is omitted.

 

it might be prudent to add 1kΩ in series with the capacitor, to limit the transient current.

Ron, you are correct. There should be a limiting resistor in my first NPN circuit. That a very large Cap! Anything from 300 Ohms to 1k should do. I'll edit that print. I don't think diodes are going to help recovery though. The base never goes any higher than a diode drop and the rev voltage from the cap isn't high enough to damage a 2N2222. It can't hurt though, so I'll add one.

 

Ron, you are correct. There should be a limiting resistor in my first NPN circuit. That a very large Cap! Anything from 300 Ohms to 1k should do. I'll edit that print. I don't think diodes are going to help recovery though. The base never goes any higher than a diode drop and the rev voltage from the cap isn't high enough to damage a 2N2222. It can't hurt though, so I'll add one.

I just meant that the diode will help discharge the cap faster on negative transitions of the input.

 

Here's an updated version using one NPN. I was able to reduce the value of C1 and greatly reduce the discharge time. The graph indicates how the output pulse width will change vs the C1 values shown. FYI, I still think a FET is preferable.

Note: I don't know why, but I find that I get a better image when I open it in a new tab vs a new window.

 

Alright, with your new way it seems to work perfectly. I used the 58K R2 version. I'm working on getting the storing values with relays working. I think the way I have it set up is going to work, I can't tell because the program I'm using says it won't work but it is really unreliable(It's basically "Baby first electronic project software"). I working on learning how to use SPICE but it's much more confusing. Also, I've realized that I can't finish this with the amount of breadboard space I have, so I'm going to have to buy a bigger breadboard.

Also I have a n-channel and a p-channel mosfet but when I ordered them(quite a while ago), I didn't really understand what they were and just hoped that I would need them later. But what I didn't notice is how big they are. They are about an inch tall and about a half inch wide. Way to big to fit in the breadboard.

Anyway, thanks for the help, and I'm sure I'll have more problems later.

 

I just meant that the diode will help discharge the cap faster on negative transitions of the input.

OK Ron, I see what you're saying and it's a good point.

Chris

 

Please note that a Diode (D2) has been added to the base of Q1 in post 23 for the reasons that Ron mentioned.

 

This is very embarrassing but after trying to get this project to work with relays I've decided to take all the relays out. Relays caused so many problems for me. I tried messing around with the circuit and replacing the relay with a transistor but I couldn't get it to work. One little change is that it needs to be the inverse of my original problem(to be perfectly clear I attached a picture of what I need).

Thanks

 

This is very embarrassing but after trying to get this project to work with relays I've decided to take all the relays out. Relays caused so many problems for me. I tried messing around with the circuit and replacing the relay with a transistor but I couldn't get it to work. One little change is that it needs to be the inverse of my original problem(to be perfectly clear I attached a picture of what I need).

Thanks

We need to know what the short output pulse is going to drive.

 

It needs to reset a SR flip-flop made with NAND gates(7438N). The amount of time that it is low doesn't matter very much, just as long as it is long enough to reset the SR flip-flop.

Also earlier in this thread I said that I didn't have any caps between 10 and 1000, but I did some searching and found a few 110uf, 220uf and 330uf caps.

 

I hope you are using a 5 volt power supply.
Are you planning to set your SR flip-flop with the input square wave?

 

I'm using 3V, the data sheet says the min is 4.5V but 3V seems to work fine. Lower voltage could just make it not work correctly or could it hurt it?

No, I'm going to use the output of a 2-bit full adder to set the SR flip-flop. The square wave will control when the full adder can set the SR flip-flop.

 

The last circuit that I gave you, that you said worked with the relay, should still work. Remove K1 and D1 and then connect the Collector of Q1 to the input of your chip.

EDIT: A pull-up resistor (1K to 3.3K) should also be connected from Q1 Collector to the 5Vcc. This will prevent the input pin on the 7438N from floating while Q1 is off. Sorry I missed that.

 

It needs to reset a SR flip-flop made with NAND gates(7438N). The amount of time that it is low doesn't matter very much, just as long as it is long enough to reset the SR flip-flop.

Also earlier in this thread I said that I didn't have any caps between 10 and 1000, but I did some searching and found a few 110uf, 220uf and 330uf caps.

It just hit me... Since your signal source is a 555 with Vcc= 5V, you should be able to use this approach using a 7404 inverter.

 

shaqywacky - Why not just use Bill Marsden's circuit, Fig. 4.3, shown in this post? http://forum.allaboutcircuits.com/showpost.php?p=117641&postcount=6

It's a proven circuit and it will do what you need. You would forget about -D1, R7, and Q1. You need the pullup R8, the PWM would go to your SR flip-flop. And you would need to set the 555 frequency to your needs.

 

@CDrive Yep, that worked. I was making it way too complicated. Thanks again.

@Shortbus Because I need both waves exactly as I drew them(IE the second wave needs to follow directly behind the rising edge of the square wave). Unless that's what that is, in which case: Thanks, but I got it working with CDrive's way.

 

@CDrive Yep, that worked. I was making it way too complicated. Thanks again.

Which one? The 7404 inverter I hope. It's been a very long time since I've used TTL, so I had to dig through my old TTL Cook Book to refresh my memory. That method was and is the accepted (SOP) way to do it.

By the way, please re-read post..
http://forum.allaboutcircuits.com/showpost.php?p=321649&postcount=36

as I edited it because there should be a pull-up resistor from the collector of Q1 (1K to 3.3K) to the 5Vcc that I neglected to mention, even if it appears to be working OK for you.