PBW needed

Yes, correct, the unpopulated holes all look to be and measure the same, both rows of them. I was unable to fully seat the 3 pin header by hand, just barely got it started. It was harder than I thought from my first attempt at lining it up for fit. Ended up having to use a small pair of flat nosed pliers to fully seat it tightly in the holes. Just have to make sure the pliers clear bottom pins while pressing it through the PCB. 

With a magnifying glass and a meter I was able to trace the 3 holes used for the pin header to their connected components and got good continuity on all three pins! Checked each pin a few times to be certain.

I was even able to remove the 3 pin 'press fit' header without damaging anything. Had to gently and very carefully pry it up with a small screw driver. As another test, I reseated it a second time (still needed pliers) and it all checked out the second time as well. These seem to work quite well, great find GRJ! Looks like we have a winner here! 

That indicates you can actually have a "no solder" option if you assemble the boards.  That should reach more people.

For pressing the header in, you might consider a small piece of wood with 3/8" hole drilled in it.  Back the PCB with the wood, line the hole up with the three holes for the header, and use a vise to gently press the header in place.  I would use my press, but if you don't have one, a vise works.  Another method is using a drill press as a press.

That is becoming an option I'll be considering, these pin headers seem to work very well. And the reduced parts needed should be an advantage as well.

I'll look into the wood block suggestion. Sounds like a good idea there too. 

You just have to insure whatever is on the bottom isn't pressed by the backing, which is why I mention a small block with just the hole for the connectors.  With only six parts including the header, it would be pretty quick to assemble these boards.  Give yourself a few bucks for the labor and make a little package that doesn't require any soldering to install in the DCS-RC. 

From some of the stuff I've seen, a lot of people would love to be able to do something like this and not have to deal with soldering.

I think that's a good idea about the no solder PBWs, I agree that lots of folks just don't want to solder stuff and would like things that just plug in. 

You were too quick replying, I was editing to add the following so I re-did it.

Need to do some testing with the new component values and check that out. I guess I'll have to rely on my Hz readings with the DMMs. The flashing LED is not synching extremely well with the second hand on my watch. It syncs for a while then gets out of step for a while. I don't have any PCBs yet so I can't try it on my layout until then. 

I think that Hantek scope may be calling me here, that would be a good way to break it in. I have been eyeing it everyday at least once.     

Once you have a decent digital 'scope for stuff like this, you'll wonder how you lived without it!   If my ATTEN ever craps out, I'll be on the Internet ordering a new one.  I'd probably go for a Rigol model, I've seen and heard good things about their products.  However, currently the ATTEN is working just fine, and it's more than sufficient for what I do here.

FWIW, I test these with a little board with a socket and an LED, I just supply it 5V with a bench supply.  I do have a frequency counter that I can hook up to check the period, but it's just easier to grab the 'scope, so that's what I normally do.

Or use the digital stopwatch function on your smartphone or traditional cellphone?  "Start" on a blink, "stop" on the 10th blink.  Should read 10.0 sec...+/- 1 sec if within 10%.

Good point about doing it for multiple resets, that does make it easier to time it without expensive equipment.  Of course, for the design verification, I'd probably still want to know exactly what several samples do by actual measurement.

gunrunnerjohn posted:

Good point about doing it for multiple resets, that does make it easier to time it without expensive equipment.  Of course, for the design verification, I'd probably still want to know exactly what several samples do by actual measurement.

How is watching the LED blink not an actual measurement? 

Stan, reaction time being what it is, it's an actual measurement colored by human reaction time.   An actual measurement, at least to me in this case, implies a piece of test equipment.

GRJ, I've heard good things about Rigol as well, very nice. I am still an apprentice and will probably have trouble using the Hantek. I saw something the other day about the Hantek 70Mhz model ($199) actually being the 200Mhz model and could be 'unlocked' with some kind of tweak or something? That was over my head, also sounded a bit questionable? Think I'll stick with the 100Mhz one we were recently looking at, the bait is working and very close to setting the hook. 

I added an LED on the breadboard from pin 3 of the 555, what I was trying to time with my watch. I am not 100% confident the DMMs I have are measuring HZ as accurately as your frequency counter would? They do seem to be in the ballpark with the 555 calculator findings though.

Stan, all my watches are analog, I'm using Mickey's hands to do the synch test.  I was trying to make use of one of the new fangled features (Hz) on some of the DMMs I have accumulated, all inexpensive of course. Good idea on the smartphone, hadn't thought of that. I'll give it a try with the LED and blink counting. 

You guys are quick!  As a friend of mine always says, "anything worth doing is worth over doing".  I would like to test this in every way I can and will probably end up learning a thing or two in the process. 

Hey, just thought I would bump this thread and see where we are at getting this mainstream. While I don't have any aversion to soldering I do like getting a complete kit to build and wanted to know if this will be offered in a completed version and one that can be a builders kit. I have built one already (thanks GRG) and due to recent changes to the layout need another. If you have un-powered yards this is the ticket to keep locos dark and silent when turning on those tracks.

I think Tom is polishing up the 555 based kit, that one will not require any calibration, just PnP.

GRJ is correct. New 555 design is working on a breadboard, but untested in actual service. I'm waiting on some PCBs to actually test it with a DCS-RC on my layout. With nothing unexpected happening, should have something ready in the next week or two.

The plan is to offer complete kits that will need assembly (soldering) and also a fully assembled unit that needs no soldering, fully PnP (thanks to the 'push-fit' pin header that GRJ found). 

All the parts and PCBs are in. I have assembled a couple and been testing them on my layout that last couple of days and have had no problems so far. My initial order was mainly for testing to see how everything went, so far so good! I have a few already spoken for, but I still have several left (kit or assembled) from the first batch. The 'Press-Fit' headers seem to be performing very well here so the assembled version can now be a completely 'no soldering involved' version.  Both versions are PnP (no adjustments needed) as mentioned earlier in the thread.

For anyone that is interested in one of these: Please send an email to my profile email address and I will email you the ordering information.

If anyone wants to make their own, attached are the revised Diptrace files and Gerber files for the newer version.

Schematic and Parts list are also included in the Instruction file (.PDF). 

Edited - 05-08-2020:  Fixed error with value of R1 on schematic and parts list.  All related files posted here have been updated.

I am getting ready to assemble one of these and I have a question about the value of R1. The part number that is listed is for a 3.0 Mohm resistor, but the diagram shows a 300 Mohm resistor.  Which value is correct?

Well, it's a 3.0mb resistor, if it were a 300mb resistor, you could just use air!

Thanks.  That's what I thought, but just making sure.

Good catch there, I've been missing that for all this time! 

The error was in the part's 'Value Description' in Diptrace. Part numbers and suppliers were all correct in the parts list.

Post above is now fixed with the correct value for R1. All the related files have also been corrected and re-posted above.

I wonder how many years between reset with a 300 meg resistor?

My boards came today from OSHPark, and everything works perfectly. Here is a picture of it connected it to the board that controls 6 of my yard tracks.

@rtr12 posted:

..

If anyone wants to make their own, attached are the revised Diptrace files and Gerber files for the newer version.

I did not realize there was much interest in the DIY version but this thread was mentioned today.  Anyway, I was looking at your parts list and found it curious the resistors are shown from Mouser while everything else is from DigiKey.  Given postage is $5 and up per order, got me wondering...

Mouser shows a minimum order of 200 pieces for the Xicon brand resistors (270-value-RC).  Earlier in this thread alternate part numbers for Vishay brand are given; and while those are available quantity 1, I was surprised to see the 3.0M resistor is a rather steep 87 cents each (47K is "only" 31 cents each) with price break at 50 pieces.

Again, from what I can tell the overwhelming majority of guys want an assembled version, but for the sake of posterity and for a DIY'er stumbling upon this in the future, perhaps a comment about the two different parts suppliers?

Both Digikey and Mouser are top-tier electronics supply houses, either is a good choice.  I think you'll find right now that there are parts shortages of many parts, including some of the parts specified on that BOM.

For the resistors, I ended up having to substitute for these at Mouser.

594-5063JD3M010F  MBA02040C3014FC100  .4watt 3.01Mohms 1%

603-MFR-12FTF52-47K  MFR-12FTF52-47K  47K OHM 1/6W 1%

Also, some of the prices have really rocketed up, the thru-hole .01uf caps are 38 cents!  They were much cheaper a few months ago!

All which goes to the point of why guys want the assembled version!

For anyone stumbling on this thread willing to, or wanting to, work at the component level, I've concluded the best way forward is to freely ask questions if something is not clear.  For example, there was a contemporaneous thread where the question was posed if the 555I version could be used instead of the 555C version (answer is yes).

Likewise, I noticed the 555CP IC chip suggested in rtr12's parts list is out of stock at DigiKey right now.

But they do have the 555CPE4 available for the same price.  Can this be substituted?  Answer: yes.  The E4 suffix has to do with "Green" (e.g., lead-free) packaging...but I'm not sure this is common knowledge to the typical OGR reader?

Right. 38 cents for a 0.01uF cap is simply too much!  The component in rtr12's BOM is 48 cents:

There are MANY suitable alternatives for less than half the cost.  But I'm not sure it's common knowledge that an X7R can be substituted for an NPO in this particular application!

And the same story goes for the resistors!  Is it obvious that a 3.01M 0.4W resistor can be substituted for a 3.0M 1/8W resistor?

There is just too much fine-print for someone who's just trying to get their DCS engine to startup silently and in command mode on a roundtable whisker or yard siding!

@stan2004 posted:

I did not realize there was much interest in the DIY version but this thread was mentioned today.  Anyway, I was looking at your parts list and found it curious the resistors are shown from Mouser while everything else is from DigiKey.  Given postage is $5 and up per order, got me wondering...

Mouser shows a minimum order of 200 pieces for the Xicon brand resistors (270-value-RC).  Earlier in this thread alternate part numbers for Vishay brand are given; and while those are available quantity 1, I was surprised to see the 3.0M resistor is a rather steep 87 cents each (47K is "only" 31 cents each) with price break at 50 pieces.

Again, from what I can tell the overwhelming majority of guys want an assembled version, but for the sake of posterity and for a DIY'er stumbling upon this in the future, perhaps a comment about the two different parts suppliers?

Until now, to my knowledge no one has tried the DIY version of the PBW that I know of? So I think you are right about there not being a lot of interest in that option. GRJ may have had some DIY inquiries, not sure about that one? They probably would have gone to him first on that one  after reading the development thread.

As for parts suppliers, at the time this list was made Digikey did not have the 1% resistors in the values needed, but Mouser had them. Seems like there was something with that cap too, but I can't say for sure there?  I believe this is what GRJ ended up with originally as well, as he was also hunting parts for the revised 555 timer version at the time. Pretty sure he also came up short at Digikey. I don't recall exactly but it seems that Mouser did not have all of the required parts either. Anyway, that's the reason for two suppliers at the time the BOM posted. Also seems like some substitutions were considered, but that didn't work out either? GRJ would probably remember that part, he has a much better memory for this stuff than I do.

It would definitely be preferable all coming from one supplier. It's an extra $4.99 for shipping at Mouser, at least it was at the time of the posting. Digikey shipping isn't like it used to be either, I think they now have a minimum similar to Mouser's. I haven't checked on these parts since the BOM was posted, but from the above it looks like GRJ may have and it sounds like he still had to get some from other suppliers. Also, I think many things seem to be in short supply these days as GRJ said above.

Yep, the old BOM didn't work, I had to scrounge around for substitutes.  Since most of the requests were for A&T, I just ordered a bunch of boards and components and assembled them.  I assemble them a panel at a time, more efficient than doing one at a time.  I have lots of the blank boards now.

I don’t know how many do try the DIY version. I just did it, a few months ago I did all the read watch to find the 1% resistors at Digi-Key. When I went to order found the TLC555CP was out of stock. When I did my quick once over figured the TLC555IP would be fine but since this popped back up figured I’d check with those smarter than me. That 10nf cap was still kinda up there but didn’t feel like finding a different one with same specs and footprint. PCB was already made. I got the 3M for 19.8¢ And the 47K .4W for 27.5¢  

I did that mainly because I didn’t know of anything I needed from mouser that I could get to justify  the cost of shipping

Any of the low power 555 DIP chips should work, the actual timing is the same for any of them.

@zhubl posted:

...

PCB was already made. I got the 3M for 19.8¢ And the 47K .4W for 27.5¢  

so just to be clear, the PCB can fit a 0.4 Watt resistor?

I didn't examine the PCB design, but the "original" BOM appears to have a 1/8W resistor (0.125 Watt).  It kind of looks like the holes are spaced 0.3" as per the 0.1" grid protoboard above.  You can generally fit a 1/4W resistor (0.25 Watt) in a 0.3" spacing.  So even better if 0.4 Watt fits in the PCB design provided.

I realize we're talking nickels-and-dimes but always good to know what options are available!

The resistor hole spacing on the the Rev. 3A version of the WD board is 6.5mm.  I picked resistors that would fit, typically 1/8w.  Some of the precision 1% resistors of higher power ratings are short enough to work also, just check the datasheet before you order them.

All good information as usual guys. The bigger power resistor did indeed just barely fit. Here’s how mine turned out. I don’t know why it never crossed my mind that I could simply get a smaller watt resistor but oh well live and learn I suppose.

The electrons will never know they have to take a detour through the big resistor.

@gunrunnerjohn posted:

The resistor hole spacing on the the Rev. 3A version of the WD board is 6.5mm.  I picked resistors that would fit, typically 1/8w.  Some of the precision 1% resistors of higher power ratings are short enough to work also, just check the datasheet before you order them.

Would there be an issue with mounting a physically "oversized" resistor to the PCB vertically (where the bottom lead goes straight down through the hole and the top lead has a ~180 degree bend to line up with the other hole?  Would this arrangement add unwanted capacitance?

It won't hurt anything Steve, but be mindful of the clearance needed between the PBW and the main board.  I'd do what Zachariah did with the resistor.  Or, you can just use the right sized resistor!