Where to get decent 120uF cap for 555

[greg123]

I'm struggling help please!

Looked on RS and Maplin and can find little 120uF which is needed for my 555 timer circuit other than low spec alloy, which I read isn't recomended for timer circuits as this circuit needs to be stable over time.

Can anyone suggest a suitable supplier for a good and stable 120uF cap for attached circuit?

Also, while on the subject I just got a basic 10mm trimmer for the Pot, it's vital that once the pot is set the circuit doesn't 'drift' much over time do I need any special kind of pot or is a normal trimmer, used once to set the time value going to be good at sitting there and not playing up/altering resistance?

Greg.

 

[Hero999]

The problem is all large capacitors have a poor tollerence, you could add a 22:muF in parallel but I don't see the point.

You could use 100:mu:F, but that would only give you a delay of 56s.

Another option would be to use 100:mu:F add and another 100k trimmer in series with the 500k trimmer which would provide a more fine adjustment.

Use a tantalum capacitor rather than an electrolytic because they have a lower leakage current.

 

[ericgibbs]

I'm struggling help please!

Looked on RS and Maplin and can find little 120uF which is needed for my 555 timer circuit other than low spec alloy, which I read isn't recomended for timer circuits as this circuit needs to be stable over time.

Can anyone suggest a suitable supplier for a good and stable 120uF cap for attached circuit?

Also, while on the subject I just got a basic 10mm trimmer for the Pot, it's vital that once the pot is set the circuit doesn't 'drift' much over time do I need any special kind of pot or is a normal trimmer, used once to set the time value going to be good at sitting there and not playing up/altering resistance?

Greg.

hi,
Get a 100uF and a 22uF 16v tantalum from Maplins and connect them in parallel.
Use a 20/22 turn Cermet potentiometer.

What accuracy in timing over time are you expecting from a 555 timer.

 

[Nigel Goodwin]

As suggested, check the tolerence on electrolytics - it's absolutely HUGE - so a 100uF and a 120uF are really the same thing.

 

[Speakerguy]

Does it have to be 555 based? A PIC 12F using internal timers/oscillator will consume less power and be way more accurate over time.

 

[Speakerguy]

Also, could you use an opamp based capacitance multiplier and a much smaller (poly film) capacitor in the 555 circuit? Digikey has .1uF to 1uF stacked film caps for not too much, add a single opamp and 3 resistors IIRC and you can easily have the equivalent of 100uF or so.

 

[greg123]

Thanks very much guys for your fast replies.

Thanks guys. 56sec max timing is absolutely fine, in fact 45 sec max would be fine so 100uf sounds like it will do the job. I couldn't find a 100uf in 16v though, as I'm running up to 15v I figured the 10v ones are not good enough. I found one here that is 20% tolerance of the Tantalum type and 47uf, if I put a pair in parallel that should do the job right?:

**broken link removed**

No it doesn't have to be a 555, just want to keep it simple as this is a first project and won't develop, it's a simple "cut the 12v trigger supply, output goes high for x seconds once only" and that's all I want it for. ;-)

 

[greg123]

Hi Eric,

I want the timer to do nothing, ever, save for when I cut 12v (9-15v actual, car circuit) to the trigger - I then want it to procuce it's high output for a period of up to 45 sec, set by the trimmer. After which the circuit needs to wait till 12v has been applied and then disconnected from the trigger for it to output again. Maximum time the circuit is powered up live is probably a few hours, during which it will probably only undergo one or two cycles as above.

Accuracy, I can set the required time with the trimmer so that's not an issue, what I do need is that once the trimmer is set, for argument's sake to give an output time of 30 seconds, that the circuit 3 years later is still producing 30 second output, give or take a second. If a year later it was only producing an output of 20 seconds, this would be a problem.

Variations in the initial min/max values of 2 or more circuits is not a problem as I say they will be individually trimmed and not used 'out of box'. It's the stability over time that I'm interested in.

I have found this trimmer, it's only 0.125w but the specs look good otherwise maybe you would be kind enough to comment if it will be okay?

https://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=1542101

Thanks for helping!

Greg.

hi,
Get a 100uF and a 22uF 16v tantalum from Maplins and connect them in parallel.
Use a 20/22 turn Cermet potentiometer.

What accuracy in timing over time are you expecting from a 555 timer.

 

[ericgibbs]

hi Greg,
Thats the type of trimmer I often use, it will be OK in your circuit.

As you are looking for longish term stability in the time period, use 1% metal film resistors in the timing section.

If its working in a car, you should consider the 'harsh' environment in a car, eg:
temperature range, humidity and vibration and electrical noise.

A low cost 9V battery will give a poor performance at very low temperatures, also use a CMOS version of the 555, it has a wider operating temperature than a standard TTL version.

Eric
PS: is Rediffusion still in Castle Boulvard.???

EDIT: if no one else posts a diagram, I'll draw and post it.

 

[greg123]

Rediffusion seems to be there according to the web, I don't know what it is but I can drive past tomorrow and check for you if you want!

Okay I have sourced resistors, they are 0.25w though but are 1% metal film. Or I can use 0.6w, but I can't get a 1k, so I would have to have a 1.1k. I think the 1k is a pulldown resistor, so if the higher wattage ones are better for reliability would a 1.1k be okay? If 0.25w is more than sufficient I'll just get a multi pack of them so I have a selection for the future.

I wasn't trying to hijack multiple threads re the circuit but I have attached it here fyi, this thread was just because I was struggling to find supplier for the cap which now seems to be sorted if I use the ebay tantalum ones 2 in parallel (47uf).

Do you recommend any web sites/software that I can 'virtual build' the circuit and test it/do diagrams? I'll search the forum later on the subject.

The circuit will be powered by car voltage so yes not too smooth and up to 15v but I have ordered the cmos 555 which is rated to 18v so that should be fine I'm thinking. 16v on the caps should still be okay with a 14.4 nominal voltage when engine is running right?

Any advice on the type/rating of diode? I can't see anything basic on RS which astounds me, but found these 1N5401 Rectifier Diode 3A 100V ones which I hope are up to the job:

**broken link removed**

So many questions... I feel a bit bad but hopefully I'll be able to help someone else one day like I do on my other forums! Greg.

hi Greg,
Thats the type of trimmer I often use, it will be OK in your circuit.

As you are looking for longish term stability in the time period, use 1% metal film resistors in the timing section.

If its working in a car, you should consider the 'harsh' environment in a car, eg:
temperature range, humidity and vibration and electrical noise.

A low cost 9V battery will give a poor performance at very low temperatures, also use a CMOS version of the 555, it has a wider operating temperature than a standard TTL version.

Eric
PS: is Rediffusion still in Castle Boulvard.???

EDIT: if no one else posts a diagram, I'll draw and post it.

 

[Rolf]

I'm struggling help please!

Looked on RS and Maplin and can find little 120uF which is needed for my 555 timer circuit other than low spec alloy, which I read isn't recomended for timer circuits as this circuit needs to be stable over time.

Can anyone suggest a suitable supplier for a good and stable 120uF cap for attached circuit?

Also, while on the subject I just got a basic 10mm trimmer for the Pot, it's vital that once the pot is set the circuit doesn't 'drift' much over time do I need any special kind of pot or is a normal trimmer, used once to set the time value going to be good at sitting there and not playing up/altering resistance?

Greg.

I have seen 555 circuits with over 10 meg for R. I would increase R to 10Meg var. + one Meg fixed (for starters). This should require just about 5uf for the timing capacitor and overall you should have a much more stable circuit.

 

[Hero999]

555 circuits or 7555 circuits?

Don't forget that as you increase the resistor value the leakage current becomes more important, past a certain point even the input current of the 555 needs to be taken into account.

Another way of increasing the delay is to add a resistor from pin 5 to +V. I can't remember the formula but it's on one of the application notes.

 

[ericgibbs]

hi Greg,

The easiest way would be to get a low power 12Vdc relay [a high resistance coil] with change over contacts.
Connect the 12V relay across the car battery via a fuse.
Use the contacts on the 12V relay to connect the junction of the 10K and 1n0 to 0V when the relay becomes denergised.
This will trigger the monostable.

It could also be done using a couple of transistors or if you have on the shelf any CMOS gates, invertors you could use an ic.

What 9V battery life are you expecting.?

Remember also that on a cold day the car battery could drop to around 9V or so when operating the starter motor.

Does this help.?

 

[greg123]

Hi Rolf, thanks for that. Is the circuit unstable as it is? Are you saying reduce the 120uf cap to 5uf and replace the 500k pot with a 10m pot?

Why would this make it better?

The power will be supplied by car +12 which will be likely 14.4v. I need the circuit to be safe from 9 to 16v just in case. Two lines of the car's 12v will be coming in, one as power and the other as a constant +ve which when switched off will switch on the timer output. Oh and though the supply is DC it isn't super smooth, though smooth enough for most stuff. I read somewhere putting a cap across the timer supply is a good idea maybe this is relevant?

Advice appreciated so I can get the bits in and start breadboarding. I know enough to bench test but have no idea on what makes a more stable circuit, all I'd be able to do is test if it works.

Greg.

I have seen 555 circuits with over 10 meg for R. I would increase R to 10Meg var. + one Meg fixed (for starters). This should require just about 5uf for the timing capacitor and overall you should have a much more stable circuit.

 

[greg123]

Hi Eric,

Won't the above circuit be triggered by loss of 12v feed into the trigger side? I think the way Hero did it that it will breifly drop to 0v across the trigger and thus when I cut the 12v supply to the trigger side (12v will be kept to the power side, naturally) it will activate the output.

I'd rather not add in another relay if I have to as in another part of the circuit (outside of the timer) I already have two and it's got to go in a small box. Hence why I was after a circuit that would do a 'delay on break' without needing additional relays/transistors if poss, KISS and all that!

So no need for 9v battery, it will be hardwired in a car!

Greg.

hi Greg,

The easiest way would be to get a low power 12Vdc relay [a high resistance coil] with change over contacts.
Connect the 12V relay across the car battery via a fuse.
Use the contacts on the 12V relay to connect the junction of the 10K and 1n0 to 0V when the relay becomes denergised.
This will trigger the monostable.

It could also be done using a couple of transistors or if you have on the shelf any CMOS gates, invertors you could use an ic.

What 9V battery life are you expecting.?

Remember also that on a cold day the car battery could drop to around 9V or so when operating the starter motor.

Does this help.?

 

[greg123]

Hey Hero,

The one I had put in my shopping basket is ICM7555IPAZ from RS here:

https://uk.rs-online.com/web/search/searchBrowseAction.html?method=getProduct&R=0308023

I hope that's compatible with the circuit above you found for me. By the way if you drew it what program did you use?

Greg.

555 circuits or 7555 circuits?

Don't forget that as you increase the resistor value the leakage current becomes more important, past a certain point even the input current of the 555 needs to be taken into account.

Another way of increasing the delay is to add a resistor from pin 5 to +V. I can't remember the formula but it's on one of the application notes.

 

[Rolf]

Hi Rolf, thanks for that. Is the circuit unstable as it is? Are you saying reduce the 120uf cap to 5uf and replace the 500k pot with a 10m pot?

Why would this make it better?

{snip}

Greg.

I don't know exactly what the temperature stability and long time accuracy of a 100uf electrolytic capacitor is. You seldom see them used in timing circuits. I would guess a 10meg trimpot will be much better especially when used with non-electrolytic capacitor.

 

[ericgibbs]

Gregg,

Won't the above circuit be triggered by loss of 12v feed into the trigger side? I think the way Hero did it that it will breifly drop to 0v across the trigger and thus when I cut the 12v supply to the trigger side (12v will be kept to the power side, naturally) it will activate the output.

The additional relay contact would only switch the trigger to the 7555 low, via the 10K/1n0/10K when the car battery is disconnected. The supply to the rest of the circuit will not be switched off.

Can you say whats the actual purpose of the application.?

eg: why do you want to detect the disconnection of the battery and whats the timer relay going to drive.? Maybe a block diagram.

 

[MRCecil]

Won't the above circuit be triggered by loss of 12v feed into the trigger side? I think the way Hero did it that it will breifly drop to 0v across the trigger and thus when I cut the 12v supply to the trigger side (12v will be kept to the power side, naturally) it will activate the output.

Hello Greg,

This will be my first post on this forum, so I hope this is received with that consideration. To your question above...

The 555, when configured in its monostable configuration as you have drawn, requires a trigger pulse to something less than 1/3 Vcc. The trigger input, pin 2, is one input of a comparator, and cannot left floating for proper operation to be viable; the high to low transition of the trigger pulse starts the timing. Simply removing the 12V will not get the job done. Pin 2 must be driven below the trigger threshold, ground is good, and back to Vcc for proper operation/timing.

In this thread you have gotten some good information. Have you looked at the 555 datasheet? In there you will find the timing formula for the monostable output pulse...t = 1.1RC, where t is the output pulse width.

If you want a 30 second output pulse, how about picking a relatively cheap alternative to two 47uf tantalums in parallel. Plug a 10uf tantalum (16v, 10%,-55to85degC, Kemet) into R= t/(1.1C) and R turns out to be 2.73M. So let Rt equal 2.49M + your 500K pot. That will give you a timing range from ~27 sec to ~33 sec. If that is too short, increase Rt. If too long, decrease Rt. R is usually cheaper than the C, and less troublesome in several respects.

You mentioned using a 1N5401 for the relay clamping diode. That is overkill to the extreme. 1N4001's are still overkill in that range of I and L kick, but common as dirt and cheap. Good luck on your project.

 

[ericgibbs]

hi MRCecil and welcome to the forum.

The 555, when configured in its monostable configuration as you have drawn, requires a trigger pulse to something less than 1/3 Vcc. The trigger input, pin 2, is one input of a comparator, and cannot left floating for proper operation to be viable; the high to low transition of the trigger pulse starts the timing. Simply removing the 12V will not get the job done. Pin 2 must be driven below the trigger threshold, ground is good, and back to Vcc for proper operation/timing.

This is the point I have been making, although the trigger is 'ac' coupled via the 1n0, the junction of the 1n0 and the input 10k must be momentarily connected to 0V.
Just disconnecting the junction, by taking off the car battery lead will not give an effective trigger pulse.