Jeff or Rodger, Can you post what you found is the cheapest way of making SSR units to go with this setup. (minius the wire to go from board to outlets) I have access to a 96 Channnel PCi card, just wondering what it would cost to add the SSR's per 4 circuits. I am assuming this design uses 4 ciruits per Rj45 jack. Might help people see also total cost to help with more sales. I am thinking of grabbing one for me.
____________________ ---> 160 LOR : 96 DIO : (2) Lamp Timers : EDM Trans
---> 25,000 up...5,000 to go
This works with the DIY SSRoz boards over at DIYChristmas, which have 4 SSR's built onto one board. You are right, each RJ45 has 4 signals (along with +5V and ground for an indicator lamp). Based on the BOM from Mouser, the cost to populate the board, is $11.85. And that's for 4 SSR's, complete with indicator LED. If you bought a large quantity, the cost per piece goes down, and you might even cover the shipping... The boards can be gotten, not sure of the cost, but will try to find out - the same gent getting the Buffer Boards made, can supply the SSROz boards, I think. Here's what the board looks like, finished:
There are several other boards out there as well, some with as many as 12 SSR's. I've been collecting "hockey puck" type SSR's, as well as I/O type from Opto22, etc. and have over 80 of them. I try to get them for about $2.00 - $2.50 each with shipping. With 192 channels, I have to keep an eye on the SSR costs, also. The thing to remember, with the ULN2803, the output is "sink", instead of source, so it switches the ground, instead of the +5. The buffer board has provision (and a necessity on the PCI_CIO96 boards!) for a separate power supply. I'm using a 5 amp supply, but I'm going to separate 2A supplies, for each board - I got a whole bunch of old CentreCom boxes, with power supplies for next to nothing, that I'm going to house the boards in/on.
Roger
Last edited on Wed Feb 4th, 2009 01:22 am by 51fordf2
Another note on that SSR board:
It fits nicely in the bottom of a 2 receptacle box.
Break the tabs on the receptacle and hook it up for a nice tight 4 channel enclosure.
The SSROz are the ones I use for this application. John Wilson is who has them made in Hong Kong. He is also the one making the DIO buffer boards. I used the DIY design of placing the SSRs into PVC tubes (pods) They worked great last year. This year, I'm putting them into the PVC "North Poles" and using them to control a bunch of leaping arches.
The DIO card is a Measurement Computing PCI-DIO96H. The "H" isn't a requirement, as the ULN chips carry the load of the LEDs and the Opto couplers. The header on the DIO board is pinned out to the cable, but the order of the channels is convoluted due to the size constraints, single sided board, and order of the pinout from the ribbon cable. It takes good notes to keep it straight when you're laying it out in the sequencer.
Anyone interested in this needs to understand that it results in strictly on/off control. There are no other lighting effects. Any lighting effect applied to the channel will result in the channel being turned on. Consider the need to build your own random pattern to include these channels in the Twinkle effect.
Jeff
Last edited on Wed Feb 4th, 2009 08:18 pm by Jeff Millard
____________________ Evan... Are you at home resting yet? I'm very glad you're feeling better!
Roger is the contact for this buy. I'm just chiming in 'cause I like to talk... He's posting it here as the "member for one year" Group Buy constraints on DIYC prevent him from doing it there. Here's the BOM for the board...
517-N2550-6002RB (50 Pin Header) 1
571-5520251-4 (Side entry RJ45) 12
660-CF1/4C471J (resistor for power LED) 1
652-4609X-1LF-470 (Resistor Net) 6
604-WP7104GC (Mini Green LEDs) 49
571-1-390261-5 (18 Pin DIP sockets) 6
511-ULN2803A (Darlington Pair DIPs) 6
Jeff
____________________ Evan... Are you at home resting yet? I'm very glad you're feeling better!
Just an update - I am waiting on the guy who is getting them made - he's in Australia, and the fires have taken a bit of his time. I'll keep all posted...
In the meantime, I decided to home-etch a couple boards, just to see how tough it would be - I left off the led's, since the plan was for these to be "disposable" once I get the other ones. I was a little worried about the one trace, for the ground as it has to go between the pins on the RJ-45 jack. I'm rather proud of how they turned out! I checked, rechecked, then checked again, before etching, but STILL made an oops - I moved 4 signals over one hole, so I etched some more, but will cut the traces, and jumper the signals in the right spots...grrrrr! Pics, for any interested - just shows that you CAN do some pretty intricate traces. I am using the magazine paper/laser printer method, with muriatic acid/peroxide mix. I'm very happy with it, $5.00 buys a whole gallon of acid, and $.50 buys a bottle of peroxide. Couple gallons for $10, instead of $10 a pint or so.
A decent etch Roger! Nice job. I really wish I didn't have so much on my plate for this year or I'd give it a try. You might want to post links to the etch how-to if you have them...
Jeff
____________________ Evan... Are you at home resting yet? I'm very glad you're feeling better!
Jeff Millard wrote: A decent etch Roger! Nice job. I really wish I didn't have so much on my plate for this year or I'd give it a try. You might want to post links to the etch how-to if you have them...
Jeff
Thanks, Jeff...I'm very happy with them.
It's actually a conglomeration of a whole bunch of different "how-to"s but here's a couple, to get someone started:
Usual disclaimers - KNOW WHAT YOU ARE DOING before attempting this at home. The chemicals are very caustic, can instantly damage certain metals, stink up the house, and inflict chemical burns. ALWAYS use the "triple A" - "Always Add Acid", which means, add the acid to the peroxide, NOT the other way around. That way, any splatter will be peroxide, not acid. Important!!
I am using 2 parts peroxide to 1 part acid. If you have an old aquarium pump laying around, if you use an air stone in the fluid, it works MUCH better/faster. Also helps keep it "fresh". The above boards took between 10 and 15 minutes to etch, while the ferric chloride took well over 1/2 hours.
(didn't see any way to make them links..is there one?? Edit: Oooops, does it automatically!!)
Roger
Last edited on Thu Feb 26th, 2009 03:17 am by 51fordf2
If anyone is interested, adjustments have been made to this board for the second production. I'm going to have another set made, and the quantity will set the price. As of now it will be around $10 plus shipping from AU per bare board. If anyone wants to get in on the run please PM me within the next week. Otherwise I'm going to place the order around next Tuesday. We will try to get a quote and estimate shipping charges before the order. I'll post a PDF of the board when I get home from work tonight.
Jeff
____________________ Evan... Are you at home resting yet? I'm very glad you're feeling better!
Jeff Millard wrote: If anyone is interested, adjustments have been made to this board for the second production. I'll post a PDF of the board when I get home from work tonight.
Jeff
Jeff, put me in for two...and I'll save you a little time - here's the pdf of the new DIO Buffer v1.1.
The error in the ULN direction has been corrected, and a shunt has been added to isolate the power source to the SSRs from the ribbon cable. I have 3 of these in service for a 96 channel, and two 24 channel boards.
Jeff
____________________ Evan... Are you at home resting yet? I'm very glad you're feeling better!
