Driving a pair of solenoids with a 555?

[EvilGenius]

Hi EG,

No that pin arrangement is intended.

Discharge gives a nice fast clean drive for the gate driver chip and the CMOS 555 Q output swings rail to rail so it generates accurate timing.

Good observation though.

spec

PS latest schematic at post #38

I just checked BOM on Mouser and Digikey. Both sites flagged IR4428 as obsolete!
Also 47uF tantalum there is a major cost difference between 5% and 10% tolerance ($2.50 vs $15.50). These things are not cheap.

 

[spec]

Hi EG

I just checked BOM on Mouser and Digikey. Both sites flagged IR4428 as obsolete!

Yes, I did know that, but I think they still available, and there are equivalents, if I remember correctly.

Also 47uF tantalum there is a major cost difference between 5% and 10% tolerance ($2.50 vs $15.50). These things are not cheap.

Hmm, I would have said a jelly bean price for a 47uF tant.

I will have a look at both components and get back.

spec

 

[spec]

Hi EG

I just checked BOM on Mouser and Digikey. Both sites flagged IR4428 as obsolete!

The IR4428 has been replaced by the IRS4428, but the IRS4428 is only available in a SOIC8 surface mount package. So go for a MicroChip TC4428A or TC4428 which are available in a DIP8 packages.

https://www.digikey.co.uk/product-detail/en/microchip-technology/TC4428AEPA/TC4428AEPA-ND/442809

As this function is an industry standard other manufacturers have functional replacements.

The TI UCC27528, UCC27328, and UCC37328 are similar but with 3A gate drive capability rather than 1.5A.

Also 47uF tantalum there is a major cost difference between 5% and 10% tolerance ($2.50 vs $15.50). These things are not cheap.

Those metal case military style tantalum capacitors are a crazy price now.

Here is a 10%, 47uF, 25V tantalum capacitor. Although it is surface mount its still a decent size so should not be too hard to fit.

A technically better capacitor would be a film type, but they are a bit pricey at £6UK and they are also a touch bulky.

spec

 

[Pommie]

Or just go for the BD139 solution. Simples.

Mike.

 

[AnalogKid]

AK's suggestion of using a digital period generator would provide a Rolls Royce plus performance but with more cost and complexity.

Don't think so, especially when relying on a tantalum capacitor for long-term stability. Here is a first pass at the RR of timing circuits. The adjustment range for the on time for either solenoid is 27 s to 63 s according to this equation (derived from the 4060 datasheet):

t = 1.1 x R x C x 16384

Note that the number of timing components is the same as with a 555.

ak

 

[EvilGenius]

Here is a board in its beginning stages. It should give you a good feel of the part sizes and how close they could possibly get. This is with SOIC-8. I can change it to DIP-8. Also you have several choices of pitch for all caps including 47u-T. I am leaning toward o.1P to compact the circuit. Board is roughly 1inch x 2 inches.

 

[spec]

Don't think so, especially when relying on a tantalum capacitor for long-term stability. Here is a first pass at the RR of timing circuits. The adjustment range for the on time for either solenoid is 27 s to 63 s according to this equation (derived from the 4060 datasheet):

t = 1.1 x R x C x 16384

Note that the number of timing components is the same as with a 555.

ak
View attachment 103691

Very neat AG- I like your timing approach.

You missed one important plus of your design: accurate and consistent 1:1 mark to space ratio.

spec

 

[AnalogKid]

I didn't miss it. I just thought the 555 crowd didn't need any more piling on.

Why does the 555 circuit have a 92:1 timing range? 1.3 s to 103 s seems a bit broad for a 2:1 requirement. A 1 M pot and 470 K fixed resistor gives a much more adjustable range.

ak

 

[AnalogKid]

Here is a board in its beginning stages.

Recommendations:
1. Open up the component spacing. This makes the board easier to assemble and much easier to troubleshoot.
2. Eliminate all component values.
3. Place a reference designator for each component somewhere where it can be seen after the board is fully assembled. Contrary to intuition, placing component values under the parts does not speed assembly or reduce errors, and makes diagnosis much more difficult.

ak

 

[EvilGenius]

I didn't miss it. I just thought the 555 crowd didn't need any more piling on.

Why does the 555 circuit have a 92:1 timing range? 1.3 s to 103 s seems a bit broad for a 2:1 requirement.

ak

A lot of great ideas in this thread. It is up to OP to prioritize cost vs. accuracy, and the range of its timing utility.
If timing is not the top priority, then I would personally go for the lowest cost and smallest PCB real estate with moderate timing accuracy. I need to break away for a few days.

 

[EvilGenius]

Recommendations:
1. Open up the component spacing. This makes the board easier to assemble and much easier to troubleshoot.
2. Eliminate all component values.
3. Place a reference designator for each component somewhere where it can be seen after the board is fully assembled. Contrary to intuition, placing component values under the parts does not speed assembly or reduce errors, and makes diagnosis much more difficult.

ak

I had to rethink how I am going to post a response. We all are professionals and have done hundreds if not thousands of circuit designs and PCP layouts. PCB design is as much an art and personal preference than it is utilitarian. If you guys are creating your schematics in Eagle, Kicad, or any other software, then you can easily convert that to a PCB layout with a rat's nest. If OP would like my help then I would gladly step up. Other than that, I am glad I came across this thread. Best of luck to you RF.
Regards,
EG

 

[spec]

I didn't miss it. I just thought the 555 crowd didn't need any more piling on.

Why does the 555 circuit have a 92:1 timing range? 1.3 s to 103 s seems a bit broad for a 2:1 requirement. A 1 M pot and 470 K fixed resistor gives a much more adjustable range.

ak

Steady on now.

Why not provide a wider range?

spec

 

[AnalogKid]

Why not provide a wider range?

Adjustability. The range in post #1 occurs over less than 30% of the 2 M pot's value, or about 78 degrees of rotation. Plus, depending on the pot being used, mechanical repeatability can be as bad as 20%. Minor component value adjustments triple the mechanical spread for the same circuit performance. Or, if you stay with the 2 M pot, you can eliminate the 22 K resistor.

ak

EDIT: never mind.

 

[spec]

Adjustability. The range in post #1 occurs over less than 30% of the 2 M pot's value, or about 78 degrees of rotation. If you stay with the 2 M pot, you can eliminate the 22 K resistor.

ak

But that is just the intention- to give a wide range of adjustment as I said previously.

spec

 

[AnalogKid]

I think we're talking past each other. Note - all of this is just opinion; no attitude expressed or implied.

An adjustment range that is a little wider that the requirements is good - it covers component tolerances. But one that is almost 50 times wider that the req can turn lots of adjustability into no *useful* adjustability at all. If the req range is a small percentage of the total range, selecting individual values (like 35 seconds vs. 40 seconds) can be very difficult (again, depending on the physical construction of the pot).

ak

 

[rfranzk]

Hello All,

AK. Thanks for the creative approach. I did request in a later post #16 that the timing adjustment range be changed to 5 seconds to 1 minute and that some indicator led's be added. Accuracy is not of prime concern on this project.

Cost is not of huge concern either so I am open to more expensive components to improve the circuits durability and accuracy.

It was not my intention to introduce any disturbance to this forum, was just looking for competent recommendations which there is certainly been a GREAT selection and participation from the members. Everyone has gone above and beyond and it is greatly appreciated.

I am just learning Eagle and would appreciate any assistance to complete this project. I am real busy at the moment and probably not able to do much for a couple of weeks.

Thanks again to all.

rfranzk.

 

[AnalogKid]

spec: I missed the req update. my bad.

ak

 

[spec]

spec: I missed the req update. my bad.

ak

No probs AK- I miss things all the time.

spec

 

[EvilGenius]

Here is a updated PCB.

 

[spec]

Hi EG,

Can you please just check that you are working with the latest issue of the schematic of post #38.

Thanks

spec