4047 monoshot(1ms pulse width)

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

4047 monoshot(1ms pulse width)

Post by vinod »

I have designed oneshot for 1ms Pulse width using 4047.

R5- makes output current >= 10mA to ensure load regulation.
C4- prevents bouncing of trigger switch.
1N4002- prevents discharging of c1 through LM317T.

1. Is it possible to see the 1ms 'short time' switching between RED & GREEN LEDs at output of 4047 ?

2. Can I use C4 in parallel with C1 as both have different roles in the circuit ?

3. 10K or 100K which one is better to pull down trigger switch to GND ?
Attachments
4047monoshot(1ms).png
4047monoshot(1ms).png (5.45 KiB) Viewed 5649 times
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

Oh wow! Here we go again. :)

As wired I don't see how C4 is going to debounce the switch.

You can see the 1ms pulse on a LED if the LED is in the off mode. However because the green LED is normally on you won't see it turn off because of persistence of vision.

I've made some changes to the circuit as I see needed.

The trigger circuit is designed to incorporate a slight delay ( about 10ms ) when activated and prevent bounce when released.

Rewired the green LED so it's also normally off. You will see both LEDs pulse on at the same time when triggered if that helps.

Steve G
Attachments
Vinod 4047 one shot.JPG
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by vinod »

sghioto wrote:Oh wow! Here we go again. :)


You can see the 1ms pulse on a LED if the LED is in the off mode. However because the green LED is normally on you won't see it turn off because of persistence of vision.


The trigger circuit is designed to incorporate a slight delay ( about 10ms ) when activated and prevent bounce when released.

Rewired the green LED so it's also normally off. You will see both LEDs pulse on at the same time when triggered if that helps.

Steve G
1. The persistence of vision at least is 1/16 Sec=62.5mS. But LED at 10th pin of 4047 turns on for 1ms only. That is I have to use oscilloscope in trigger acquisition mode with proper trigger level and trigger slope for detection of 4047 1ms output instead of LEDs.

2. How do you calculate 10mS ? Is it discharging time of C4 through switch ? (When bouncing occurs at switch contacts , how can C4 discharge through the switch !!)

If that is correct my interpretation is that, during switch press C4 would be charged through 100K and 100R to 5V and this time is very high as compared to discharging interval. Due to instant release of switch, C4 will be in partially charged condition. When switch is released C4 will start discharging through the switch thus opposing bounces at the 8th pin of 4047. Is it OK ?

Then I can avoid this partial charging by making the changes as shown in the below attachment.

Another debouncing arrangement is also shown.

3. The output source current of HCF4047 at 5V is 1mA. Is it enough to drive output LEDs ?
Attachments
Cd4047Debounce.jpg
4047monoshot1(1ms).png
4047monoshot1(1ms).png (5.34 KiB) Viewed 5638 times
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

3. The output source current of HCF4047 at 5V is 1mA. Is it enough to drive output LEDs ?
Actually No. You need to use a buffer or transistors.
The persistence of vision at least is 1/16 Sec=62.5mS. But LED at 10th pin of 4047 turns on for 1ms only.
As I said, you can see a LED pulse on even though for only 1 ms. I verified this myself but you will need to use a bright LED drawing at least 10ma.
my interpretation is that, during switch press C4 would be charged through 100K and 100R to 5V
That is correct. The 10ms delay is one time constant ( .1uf X 100K = .01s ). As soon as the trigger switch contact opens C4 will start charging to 5 volts mainly through R8. C4 is initially a short when the switch first opens so any bounce at this time is ignored. The actual delay will depend on the minimum value of input voltage required to trigger the 4047. At 5 volts Vcc the typical value is 2.25 volts to trigger at 25 degrees celcius. So it will take C4 apprx 7ms to charge to that value. When the switch is released nothing happens until the NC contact makes closure. At this point C4 will rapidly discharge through the 100 ohm resistor and the switch. Any bounce at this time is ignored as C4 tries to recharge. Remember a lot of this depends on the switch itself and how much bounce is actually happening and the amount of time required to go from NC to NO. In your modified circuit C4 will charge instantly to 5 volts and trigger the 4047. If the switch bounces the C4-R8 combo will hold the 5 volts for about 1ms. If the bounce is over 3 ms the 4047 will trigger at least twice. That's why I put the delay in. But again it depends on the switch which is why we discussed the bounceless switch using the inverters. What happened to that idea?


Steve G
Attachments
Vinod 4047 one shot.JPG
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by vinod »

sghioto wrote: The actual delay will depend on the minimum value of input voltage required to trigger the 4047. At 5 volts Vcc the typical value is 2.25 volts to trigger at 25 degrees celcius. So it will take C4 apprx 7ms to charge to that value.

Remember a lot of this depends on the switch itself and how much bounce is actually happening and the amount of time required to go from NC to NO.


Steve G
For HCF4047BE, trigger high voltage should be greater than or equal to 3.5V and low level should be less than or equal to 1.5V.
Your arrangement will take around 12mS to charge to 3.5V. It means I have to push the switch for minimum 12mS time to trigger the 4047.

Can I reduce this time by changing 100K to 10K ?
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

vinod said:
It means I have to push the switch for minimum 12mS time to trigger the 4047.
Yes! But is that a concern and if so why? 12ms is not much time to wait when pushing a switch manually.
Can I reduce this time by changing 100K to 10K ?
You can but it will defeat the purpose of the debounce.

Steve G
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

output drive with debouncer

Post by vinod »

Here is another arrangement with output drive and debouncer.
Attachments
4047_324.png
4047_324.png (6.04 KiB) Viewed 5630 times
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

Looks Good to me. :)

Now's the time to build the circuit and see if it works as designed.

Steve G
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by vinod »

sghioto wrote:
Now's the time to build the circuit and see if it works as designed.

Steve G
As per my friends opinion the debounce circuit has a slow rise time and fast fall time, so would best be used for triggering the -T rather than the +T pin of the mono.


Hence I simply change the trigger line from 8th pin to 6th pin (no changes in other parts).

OR

Can I put two schmitt NOT gates in series between trigger line and debounce output so that sharp edges trigger signals can generate ?. How to take care Lower and Upper threshold of schmitt trigger NOT during the design of debouncer ?
Attachments
4047_324_3904.png
4047_324_3904.png (5.95 KiB) Viewed 5626 times
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

Hence I simply change the trigger line from 8th pin to 6th pin (no changes in other parts).
OK but I would reverse the connections on the switch. A pulse will generate only when the trigger is released unless that's OK.
Can I put two schmitt NOT gates in series between trigger line and debounce output so that sharp edges trigger signals can generate
No I wouldn't do that. If you are going to add circuitry then use the two stage inverter debouncer circuit we discussed on your previous post. This provides clean output with less parts. The pulse duration does not affect the output pulse in the 4047 monostable

Steve G
Attachments
Vinod 4047 one shot.JPG
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

-ve edge triggering during switch pushing

Post by vinod »

sghioto wrote:
OK but I would reverse the connections on the switch. A pulse will generate only when the trigger is released unless that's OK.

Steve G
Do you mean this type of arrangement so that triggering happens when the switch is pressed not 'released' ? :idea:
It has an added advantage that the minimum time required for keeping the switch pressed is also get reduced. :grin:
Attachments
4047_324_3904(2).png
4047_324_3904(2).png (5.98 KiB) Viewed 5621 times
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

Do you mean this type of arrangement so that triggering happens when the switch is pressed not 'released' ?
Yes!

Steve G
User avatar
vinod
Posts: 148
Joined: Fri Feb 12, 2010 12:37 pm
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by vinod »

sghioto wrote: Yes!

Steve G
I'm afraid that the very low discharge time(0.01ms) will lose the function of debounce(as bounces lasts for some time after releasing the switch- from NO to NC).
Can I solve the problem by swapping 100 Ohm & 100K ?

But it will increases the minimum time required for keeping the switch pressed.
@VinodQuilon
sghioto
Posts: 326
Joined: Wed Dec 26, 2007 5:16 pm
Location: Colorado
Contact:

Re: 4047 monoshot(1ms pulse width)

Post by sghioto »

I'm afraid that the very low discharge time(0.01ms) will lose the function of debounce(as bounces lasts for some time after releasing the switch- from NO to NC).
Can I solve the problem by swapping 100 Ohm & 100K ?
Don't change the resistor value. You're thinking about it wrong. You want a fast discharge and slow rise time.
Most of the bounce occurs when the switch makes contact at the NO contact. At this point the -trigger input is grounded and any bounce is suppressed by C4 as it tries to charge up through R8. Same thing happens when the switch is released. Any bounce at the NC contact doesn't matter.
Steve G
Post Reply

Who is online

Users browsing this forum: Ahrefs [Bot], Amazon [Bot] and 36 guests