I have a circuit attached to post #37

Question ... Is the R5 10K PULL-UP resistor go infront or behind the 1N4148 diode? .. Or does it NOT matter.

Sorry about that. I am getting old and blind.

Its made properly.

 

I believe I know the answer to this. but would like confirmation. ... My whole bread board is based on using the DIP CD4538BE chips ... But .. I want to use SOIC−16 SMD 4538 chips to conserve on space. The question I am running into when looking on mouser for parts is ... What about the IC voltage handling. Most of the SOIC−16 4538 I have clicked on have a max voltage rating of 7VDC. But I need 12VDC. ... according to the datasheet, I believe I have found a 12VDC chip ... https://www.mouser.com/datasheet/2/308/MC14538B-D-77815.pdf

Am I correct in thinking the MC14538BDR2G is ok for 12VDC?

TONY

 

Most of the SOIC−16 4538 I have clicked on have a max voltage rating of 7VDC.

The package has nothing to do with the supply voltage. If it's CD4538 or equivalent, e.g. MC14538, it will tolerate a supply voltage of 3-18V.

However, be aware that there can be significant voltage spikes to automotive electronics. There was a recent thread about devices being damaged. Some good protection info was given.

 

Hi
The 74xxxx series supply max is 7v.

I would recommend the CD14538B because it has dual Schmitt trigger inputs. Supply range is 3-18v. It is also intended to replace the CD4538B.
The SOIC package is CD14538BM.

eT

 

However, be aware that there can be significant voltage spikes to automotive electronics. There was a recent thread about devices being damaged. Some good protection info was given

I didn't know about that automotive spike post. But I was going to add the 100uf cap for some supply smoothing and the zenor, per eetech00 suggestion.

The 74xxxx series supply max is 7v.

That the detailed info I overlooked. Thanks!

I would recommend the CD14538B

I will go this route. Thanks.

TONY

 

I don't know what the diode at the input of Q1A is for(?)

I thought it would be necessary because of the pull-up resistor. If not needed, I will omit. I did notice that the diode in question (Your D4, my D1) were backward from each other. Is my cathode facing incorrectly?

a 0.047u cap for some contact debouncing.

With a Pull-up resistor. I did NOT think this would be necessary. Thank you.

Low power LED indicators (2V,2ma).

These LEDs will only illuminate when the 4538 emits an output, correct? .. Are these necessary?

Thank you
TONY

 

I thought it would be necessary because of the pull-up resistor. If not needed, I will omit. I did notice that the diode in question (Your D4, my D1) were backward from each other. Is my cathode facing incorrectly?

The way I show it is correct...but it’s not needed.

These LEDs will only illuminate when the 4538 emits an output, correct? .. Are these necessary?

Only if you want them. The output of the IC is limited to 6mA max. The modfet gate doesn’t draw any appreciable amount of current. Notice..these are N channel mosfets.

 

With a Pull-up resistor. I did NOT think this would be necessary. Thank you.

Typo...should be 0.47uf. This is only needed if the sensor output is a mechanical contact of some type.

 

One item I missed tho is there should be a diode across each timing resistor...cathode to +v and anode to junction of timing resistor and timing capacitor.

 

Typo...should be 0.47uf. This is only needed if the sensor output is a mechanical contact of some type.

I saw the .47uF on your drawing and went with that. I also saw that all the caps in your drawing were indicated as ceramic. I am using tantalum on everything. Is that an issue?

One item I missed tho is there should be a diode across each timing resistor...cathode to +v and anode to junction of timing resistor and timing capacitor.

Thanks.I never would've known that. 1n4148 ok?

Tony

 

I never would've known that. 1n4148 ok?

I mentioned the diode to protect the protection diodes in post #31.

Just about any diode should work; lower Vf is better. Germanium or schottky would be certain to protect the timer, but I'd consider that overkill.

 

I mentioned the diode to protect the protection diodes in post #31.

Was that the datasheet diagram. diode indicated as Dx ... I didn't pick-up on that.

Thank you
TONY

 

Just about any diode should work; lower Vf is better. Germanium or schottky would be certain to protect the timer,

Looks like a Power Schottky Low VF STPS0520Z would could work.

but I'd consider that overkill.

I leave room for it

Thanks
TONY

 

Any schottky diode would work. They cost about 10x more than a signal diode, but still relatively inexpensive.

 

I saw the .47uF on your drawing and went with that. I also saw that all the caps in your drawing were indicated as ceramic. I am using tantalum on everything. Is that an issue?
Thanks.I never would've known that. 1n4148 ok?

Tony

The 1N4148 diodes shown is not a good choice. Use 1N4004 or equivalent.
I'd use 1N4004 for all diodes (except the Zener of course). They don't have to be Schottky diodes.​

Zener diode max current should be at least 1A.

Bypass caps can be ceramic caps. Timing caps should be tantalum. All caps 25v minimum.

I'll post an updated schematic.

 

Hi

Schematic update.

Added power down diodes.
Changed all diodes to 1N4004.
Added a POR circuit just in case its needed.

eT

 

Thank you for the update schematic.

Zener diode max current should be at least 1A.

Did you mean 1 amp or .1 amp? ... Using Mouser on-line part selection, I could NOT find a 1 amp 18V Zener. 90v or 100v is the lowest voltage I could select with 1 amp Zener current.

I did noticed that on your diagram. there is a difference between my post #37 which I tested on my breadboard. And your circuit in post #40.
In my circuit #37. U1b pin 9 is going to U2a pin 4.

When I first set up my breadboard. I did originally place U1b pin 9 going to U2a pin 5, as stated by dl324 in post #29. But the circuit did not work until I moved to U2a pin 4.

I need to buy more parts to breadboard your post #40 circuit ...

Question ... can I still use PNP transistors instead of mosfets? My local electronic shop does NOT carry those mosfets for me to test on my breadboard. But I do have the transistors.

Thank you

TONY

 

Did you mean 1 amp or .1 amp? ... Using Mouser on-line part selection, I could NOT find a 1 amp 18V Zener. 90v or 100v is the lowest voltage I could select with 1 amp Zener current.

You should use a TVS diode instead of a zener and put a fuse upstream of the diode.

Zener diodes can fail open circuit, TVS diodes are designed to fail shorted (unless they explode). If the TVS diode draws significant current, a fuse can protect it.

When I first set up my breadboard. I did originally place U1b pin 9 going to U2a pin 5, as stated by dl324 in post #29. But the circuit did not work until I moved to U2a pin 4.

I'll look at this later...

I need to buy more parts to breadboard your post #40 circuit

Don't use rectifier diodes; the forward voltage drop is too high to protect the diode in the one shots.

At 0.1A, 1N4004 will have a forward voltage of 0.8V and the internal diode will be close to 0.625; so the rectifier diode won't offer much protection.


A signal diode like 1N4148 would be a better choice:


One of these schottky diodes will offer even better protection:

Question ... can I still use PNP transistors instead of mosfets? My local electronic shop does NOT carry those mosfets for me to test on my breadboard. But I do have the transistors.

PNP transistors should be fine. Don't think you ever mentioned the current needed, but it should be low if you're just shorting a switch input.

 

Thank you for the update schematic.
Did you mean 1 amp or .1 amp? ... Using Mouser on-line part selection, I could NOT find a 1 amp 18V Zener. 90v or 100v is the lowest voltage I could select with 1 amp Zener current.

What you want is a reasonably robust Zener diode that can withstand a voltage surge.
A 18v 1W Zener would be good. They cost about 0.25 cent USD.

I did noticed that on your diagram. there is a difference between my post #37 which I tested on my breadboard. And your circuit in post #40.
In my circuit #37. U1b pin 9 is going to U2a pin 4.

When I first set up my breadboard. I did originally place U1b pin 9 going to U2a pin 5, as stated by dl324 in post #29. But the circuit did not work until I moved to U2a pin 4.

U2A should be triggered at the end of U1B timing period. So U2A will need to be triggered by the falling edge of the output of U1B.
That means U1B-10 (Q) should connect to U2A-5 (trailing edge trigger).

I need to buy more parts to breadboard your post #40 circuit ...

Question ... can I still use PNP transistors instead of mosfets? My local electronic shop does NOT carry those mosfets for me to test on my breadboard. But I do have the transistors

Thank you

TONY

NPN transistors would be more appropriate because your are switching the MCU input to ground.
But you might get away with PNP. Connect the PNP emitter to +V, collector thru a 10k resistor to ground, base thru 10k to Qbar.
The junction of the PNP collector and resistor connects to the MCU input.
The issue with a NPN or PNP transistor is they are current driven, so would draw about 2-3 millamps from the IC output. Mosfets are voltage driven and draw very little current thru the gate. The CD14538 can output 6ma max so with LEDs it approaching the limit. You would need to allow the PNP transistor to drive the Leds as well.

eT

 

You should use a TVS diode instead of a zener and put a fuse upstream of the diode.

Zener diodes can fail open circuit, TVS diodes are designed to fail shorted (unless they explode). If the TVS diode draws significant current, a fuse can protect it.
I'll look at this later...

Not disagreeing, but a TVS would be more expensive. A Zener should be good enough for this.

Don't use rectifier diodes; the forward voltage drop is too high to protect the diode in the one shots.

At 0.1A, 1N4004 will have a forward voltage of 0.8V and the internal diode will be close to 0.625; so the rectifier diode won't offer much protection.
View attachment 155663

A signal diode like 1N4148 would be a better choice:
View attachment 155664

One of these schottky diodes will offer even better protection:
View attachment 155665
PNP transistors should be fine. Don't think you ever mentioned the current needed, but it should be low if you're just shorting a switch input.

I used 1N4004 diodes in a surge capacity for years with no real issues. But 1N4148 can be used as a option except for the reverse polarity protection. Should be 1N4004 or better robust diode.