• For a Limited Time the R2R Partner Membership is NOW OPEN! Get some cool swag and chances to win part of over $20,000 in prizes! Click here for more details

Reef-Pi Controllers: New all-in-one with power bar, and the original Pico

OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
Sensor cables remain such a huge bogeyman.

So, I bought another set of 20 sensor cables. Same supplier and same seller (Drok). Physically, they look the same.

Start powering them up....

1600975331018.png


Low and behold, a fake. It also doesn't work on parasitic power at all like the earlier batch of cables did. And the out of box temperature performance is out of spec (not 0.5C).

Is it too much to ask for to have a product be real? :(
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
Stuffed those sensors back into the box they came in, and now back at figuring out my "other cabling" problems as we do some R&D on the power strip controller concept.

First off is the connection point for sensors to the controller. Previously, I was standardizing on

- TRS stereo connectors for 1-wire temp devices
- Phoenix Combicon PTSM poke-in wire terminal blocks (pluggable and fixed). These work great for LEDs, but they were a pain to use with a variety of random cables people like to poke in.

Other controllers (like @robsworld78 ) make use of Buchanan pluggable terminal blocks. Like the PTSM, just with a screw instead of poke in. Also a solid choice - millions of industrial devices use them.

I'd like to "fall back" and pick a single type of connector, and I'm tempted to use JST "XA" series of connectors. Its a crimped terminal style, so its much easier to pre-fab cables for, it has the distinct advantage of being a board mount and in-line variants (extenders, etc). And most importantly, its a locking connector, e.g. its very resistant to pulling out of the socket, and the sockets are reinforced.

1601001966834.png


I've been playing around with the connector on-board and as wires, and think its pretty manageable, and not too small and flimsy. It doesn't have the appeal of a fully over-molded connector, but lets be honest there is no way thats in budget..

Here I am mocking up a float sensor assembly - the float is (missing connectors since I didn't buy a 2 pin yet) disconnectable from the extender cable, which allows you to get the mounting nut over the wire. The 3-pin main cable is plugged into a board socket. All crimping using the official crimp tools so nothing weird and bent. If you wanted to DIY cables, you can even buy pre-crimped wire leads of various lengths.


IMG_7308.jpg


On the controller end it would look like this:

1601002280101.png


Some astute readers will notice the little cute mono OLED panel, and that the connectors are all three pin instead of two.

I have mocked up switchable sensor input circuit under software control to provide both a simple contact open/close sense, and full 3/5/12V power and 1-wire communication support for complex sensors. By making sure each connector is isolated, a single bad 1-wire sensor won't bring down the other sensor ports. This is me taking the prior PicoMod "plug in card" concept, and slamming it into a cable (with one less pin over I2C!)

I'm still working on a good source of temperature probes - that part is super annoying.

And finally, I'm testing cable layouts and connectors for the power inlet jack. I'm using quick connect wire leads from the inlet port since they're more durable, and allow for the use of a CT sense transformer (vs a soldered to board inlet, and less poking erratically inside a small space for solar tab inlets). The inlet is pre-fused, so all the wiring has to run through that before going anywhere else. Once again, crimping with the official ratcheting Molex crimp tools for each style of contact to make sure the pressure and shape is correct. I probably own more in crimp tools than my old commuter car is worth :scream:



IMG_7309.jpg


Looking forward to getting the relay boards in and installed before milling out the case "cap" and designing end-caps.
 

Bigtrout

Valuable Member
View Badges
Joined
Dec 16, 2018
Messages
1,159
Reaction score
2,777
Sensor cables remain such a huge bogeyman.

So, I bought another set of 20 sensor cables. Same supplier and same seller (Drok). Physically, they look the same.

Start powering them up....

1600975331018.png


Low and behold, a fake. It also doesn't work on parasitic power at all like the earlier batch of cables did. And the out of box temperature performance is out of spec (not 0.5C).

Is it too much to ask for to have a product be real? :(
The supplier may have no idea they have fake chips, seems fakes are everywhere. 95 percent real chips may be the best you find on amazon or fleabay!
Its not only chips that get faked, but just about everything in the supply chain. Had a big issue not too long ago in my area where Square D QO breakers(arguably the best breakers available for reaidential use) were being faked and one rather large electrical supplier got caught selling them by the thousands.
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
The supplier may have no idea they have fake chips, seems fakes are everywhere. 95 percent real chips may be the best you find on amazon or fleabay!
Its not only chips that get faked, but just about everything in the supply chain. Had a big issue not too long ago in my area where Square D QO breakers(arguably the best breakers available for reaidential use) were being faked and one rather large electrical supplier got caught selling them by the thousands.
I can totally believe it. As you said even the distributors sometimes get (intentionally or unintentionally) fake products into their supply chain. There are so many DS18B20 clones in China that keeping stock straight is probably impossible unless you know exactly what to test, but at the price for the Drok cables by far most of the price goes into the actual wire leads, and not in the sensor end (copper is expensive, small semiconductors are basically free these days).
 

Ranjib

7500 Club Member
View Badges
Joined
Apr 16, 2016
Messages
8,286
Reaction score
14,758
Location
Palo Alto, Bay Area
Sensor cables remain such a huge bogeyman.

So, I bought another set of 20 sensor cables. Same supplier and same seller (Drok). Physically, they look the same.

Start powering them up....

1600975331018.png


Low and behold, a fake. It also doesn't work on parasitic power at all like the earlier batch of cables did. And the out of box temperature performance is out of spec (not 0.5C).

Is it too much to ask for to have a product be real? :(
:-( , I can’t believe even drok had fake sensors. I think they don’t realize . I’m very curious how come it’s only a few.
Is there any US based manufacturer?
 

Ranjib

7500 Club Member
View Badges
Joined
Apr 16, 2016
Messages
8,286
Reaction score
14,758
Location
Palo Alto, Bay Area
Stuffed those sensors back into the box they came in, and now back at figuring out my "other cabling" problems as we do some R&D on the power strip controller concept.

First off is the connection point for sensors to the controller. Previously, I was standardizing on

- TRS stereo connectors for 1-wire temp devices
- Phoenix Combicon PTSM poke-in wire terminal blocks (pluggable and fixed). These work great for LEDs, but they were a pain to use with a variety of random cables people like to poke in.

Other controllers (like @robsworld78 ) make use of Buchanan pluggable terminal blocks. Like the PTSM, just with a screw instead of poke in. Also a solid choice - millions of industrial devices use them.

I'd like to "fall back" and pick a single type of connector, and I'm tempted to use JST "XA" series of connectors. Its a crimped terminal style, so its much easier to pre-fab cables for, it has the distinct advantage of being a board mount and in-line variants (extenders, etc). And most importantly, its a locking connector, e.g. its very resistant to pulling out of the socket, and the sockets are reinforced.

1601001966834.png


I've been playing around with the connector on-board and as wires, and think its pretty manageable, and not too small and flimsy. It doesn't have the appeal of a fully over-molded connector, but lets be honest there is no way thats in budget..

Here I am mocking up a float sensor assembly - the float is (missing connectors since I didn't buy a 2 pin yet) disconnectable from the extender cable, which allows you to get the mounting nut over the wire. The 3-pin main cable is plugged into a board socket. All crimping using the official crimp tools so nothing weird and bent. If you wanted to DIY cables, you can even buy pre-crimped wire leads of various lengths.


IMG_7308.jpg


On the controller end it would look like this:

1601002280101.png


Some astute readers will notice the little cute mono OLED panel, and that the connectors are all three pin instead of two.

I have mocked up switchable sensor input circuit under software control to provide both a simple contact open/close sense, and full 3/5/12V power and 1-wire communication support for complex sensors. By making sure each connector is isolated, a single bad 1-wire sensor won't bring down the other sensor ports. This is me taking the prior PicoMod "plug in card" concept, and slamming it into a cable (with one less pin over I2C!)

I'm still working on a good source of temperature probes - that part is super annoying.

And finally, I'm testing cable layouts and connectors for the power inlet jack. I'm using quick connect wire leads from the inlet port since they're more durable, and allow for the use of a CT sense transformer (vs a soldered to board inlet, and less poking erratically inside a small space for solar tab inlets). The inlet is pre-fused, so all the wiring has to run through that before going anywhere else. Once again, crimping with the official ratcheting Molex crimp tools for each style of contact to make sure the pressure and shape is correct. I probably own more in crimp tools than my old commuter car is worth :scream:



IMG_7309.jpg


Looking forward to getting the relay boards in and installed before milling out the case "cap" and designing end-caps.
This is going to be epic. If done right , this will become a formidable Turn key solution .
I’m still wrapping my head around. This power strip will connected to pi with USB? And there will be a new driver for this ?
 

mnl

New Member
View Badges
Joined
May 20, 2019
Messages
15
Reaction score
23
The devices can be sourced from at least Mouser and Digi-Key. They are the real Maxim part, about $5 each, $4 in quantity 25. You would need to attach leads and pot them. The general rule of thumb is anything not sourced from a recognized distributor will ne counterfeit, or at least I assume that.
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
This is going to be epic. If done right , this will become a formidable Turn key solution .
I’m still wrapping my head around. This power strip will connected to pi with USB? And there will be a new driver for this ?

Yes. I'm going to kick start it with the File driver (and a bunch of watched files sitting in /var/run), and we can work out more direct integration.

I'm planning on supporting "back feeding" power to the Pi through the same USB connection so no extra supply is needed. Insert the Raspberry Pi in the case of your choosing (or not!).
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
The devices can be sourced from at least Mouser and Digi-Key. They are the real Maxim part, about $5 each, $4 in quantity 25. You would need to attach leads and pot them. The general rule of thumb is anything not sourced from a recognized distributor will ne counterfeit, or at least I assume that.
I'm resorting to just this, except to save my sanity its a small panel for the MSOP package parts. I'm still working on what to pot them into :)
 

Bigtrout

Valuable Member
View Badges
Joined
Dec 16, 2018
Messages
1,159
Reaction score
2,777
I'm resorting to just this, except to save my sanity its a small panel for the MSOP package parts. I'm still working on what to pot them into :)
Yes it would have to conduct heat fairly well but not electricity. They do make thermally conductive epoxies with non electrically conductive fillers like boron nitride. Have no idea on the cost of such things. Polytec TC437 looks perfect. Room temp cure, made specifically for temp sensors. Cost and availability may be the killer though.

Perhaps some type of thin saltwater safe platic tube filled with silica sand and the end epoxy sealed.

Just spitballing here
 

robsworld78

Active Member
View Badges
Joined
Feb 14, 2020
Messages
218
Reaction score
474
Location
Edmonton, Canada
1600751409347.png


Just a sneak peak of the assembly. There are a few components I'm still sizing mechanically, such as strain relief for the outlets to the top case, and ground wire bonding. I still need to work on the logic board side of this, but so far we have:

- USB out to a Raspberry Pi including powering. The strip has a 20W supply which is plenty for some sensors, the relays, and a RPI4. Something like a doser module will require a second supply - I'd like to include a bit more but space is at a premium.
- Relays are TV8 rated, and designed for large inductive load switching with suitable suppression to prevent relays from both leaking at low load and arcing which can weld them closed over time.
- There is a current transformer (can't see it in the picture) which provides current monitoring of the whole strip.
- Planning on USB, a serial CAN bus (for more strips and modules), a temperature probe 1-wire port, as well as a few water level connectors. Nothing extra fancy, but from a "basic aquarium function" standpoint it gives the bulk of whats needed.

As I mentioned, prototype boards are ordered. Just don't have them in hand yet.

The mechanical design for the testing load is also coming along nicely (bonus pictures). No you won't be able to have one (unless you really do want an 2000W rack mount space heater..., its not cheap since the resistors alone run $600). Heatsinks are 4x 12" long segments of the HeatsinkUSA 4.6" extrusion, combined with 120mm fans pushing 300-400cfm each.

1600752648253.png
Looking really good @theatrus, I can see you're going all out, looking forward to seeing the finished product. So of course I have to ask, how much?

When you are done with your load tester feel free to send it my way so I can recycle it and I won't even charge you because I care. :)

I really like the CAN bus, based on the little I know that's the proper way to communicate, heck if it's good enough for Airbus it should serve an aquarium controller well, great marketing ploy too. :)

The connectors you choose are better than what I'm using but won't be fun making cables, good luck with that. That's why I went with the pluggable screw terminals, anyone can use them but yeah they can cut thin wire and it's hard to spot.
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
Yes it would have to conduct heat fairly well but not electricity. They do make thermally conductive epoxies with non electrically conductive fillers like boron nitride. Have no idea on the cost of such things. Polytec TC437 looks perfect. Room temp cure, made specifically for temp sensors. Cost and availability may be the killer though.

Perhaps some type of thin saltwater safe platic tube filled with silica sand and the end epoxy sealed.

Just spitballing here
Yeah, we're thinking alike. In fact I just tossed a panel of these together and plan to go hunting for tube material.

IMG_7314.jpg


Because I can bulk load the little MSOPs in the pick and place, it saves having to solder to wires, heatshrink, etc the TO92 packaged ones. Still need to stick down the three terminals to the adapter, but its less steps.
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
Yeah, we're thinking alike. In fact I just tossed a panel of these together and plan to go hunting for tube material.

IMG_7314.jpg


Because I can bulk load the little MSOPs in the pick and place, it saves having to solder to wires, heatshrink, etc the TO92 packaged ones. Still need to stick down the three terminals to the adapter, but its less steps.
Past me designed these to "just" fit into a 1/4" ID section of tubing. Thanks past me, since I just picked up some sample materials from McMaster to test out what to do. Note that making the PCBs any narrower is effectively at limits for most PCB V-scoring machines, so not possible. The PCB thickness here is also at minimum to avoid more thermal mass - its about the thickness of a light piece of card stock.

I also need to test gluing end caps to the bottom of the plastic tube - is a rough band saw cut enough, or do I need to use the tabletop router to get a very smooth finish? The potting compound will fully seal the sensors, but I don't want the potting compound leaking out when I pour it in :)
 

Bigtrout

Valuable Member
View Badges
Joined
Dec 16, 2018
Messages
1,159
Reaction score
2,777
Past me designed these to "just" fit into a 1/4" ID section of tubing. Thanks past me, since I just picked up some sample materials from McMaster to test out what to do. Note that making the PCBs any narrower is effectively at limits for most PCB V-scoring machines, so not possible. The PCB thickness here is also at minimum to avoid more thermal mass - its about the thickness of a light piece of card stock.

I also need to test gluing end caps to the bottom of the plastic tube - is a rough band saw cut enough, or do I need to use the tabletop router to get a very smooth finish? The potting compound will fully seal the sensors, but I don't want the potting compound leaking out when I pour it in :)
Fine blade bandsaw cut should be ok, plastic cuts pretty clean with a fine tooth metal cutting blade, and a quick hit with a little sandpaper should be good enough to glue the caps on...
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
Looking really good @theatrus, I can see you're going all out, looking forward to seeing the finished product. So of course I have to ask, how much?

When you are done with your load tester feel free to send it my way so I can recycle it and I won't even charge you because I care. :)

I really like the CAN bus, based on the little I know that's the proper way to communicate, heck if it's good enough for Airbus it should serve an aquarium controller well, great marketing ploy too. :)

The connectors you choose are better than what I'm using but won't be fun making cables, good luck with that. That's why I went with the pluggable screw terminals, anyone can use them but yeah they can cut thin wire and it's hard to spot.
TBD, but absolutely staying under the 199 point - the variance will be how much "extra" can come with it (sensor? power cord? free sticker?). I'm still sorting out the logistics of machining the case and material for the end caps (e.g., I'd much prefer aluminum, but way more than an FR-4 end cap) which right now is the biggest variance in cost outside of assembly time.

Happy to send you a load or even some extra PCBs for the load components (seriously, I'll have extras). I'll need it for ongoing test, along side the other big tooling investment, the HiPot tester to make sure all the supposed insulation and grounding is correct for every unit out the door. Also a pretty dangerous piece of kit since every unit will be run up to about 1.3kV (the tester can go to 5kV but thats mostly for doubly insulated ground-less designs at higher voltages, but I'm not going to pretend that its not fun to think about). It gets its own test bench setup, since at those voltages you can't test on ESD benches ;)

hipot.jpg
 
Last edited:

mnl

New Member
View Badges
Joined
May 20, 2019
Messages
15
Reaction score
23
Fine blade bandsaw cut should be ok, plastic cuts pretty clean with a fine tooth metal cutting blade, and a quick hit with a little sandpaper should be good enough to glue the caps on...
Most potting epoxy is about the viscosity of water, easy enough to use that to check.
 

robsworld78

Active Member
View Badges
Joined
Feb 14, 2020
Messages
218
Reaction score
474
Location
Edmonton, Canada
TBD, but absolutely staying under the 199 point - the variance will be how much "extra" can come with it (sensor? power cord? free sticker?). I'm still sorting out the logistics of machining the case and material for the end caps (e.g., I'd much prefer aluminum, but way more than an FR-4 end cap) which right now is the biggest variance in cost outside of assembly time.

Happy to send you a load or even some extra PCBs for the load components (seriously, I'll have extras). I'll need it for ongoing test, along side the other big tooling investment, the HiPot tester to make sure all the supposed insulation and grounding is correct for every unit out the door. Also a pretty dangerous piece of kit since every unit will be run up to about 1.3kV (the tester can go to 5kV but thats mostly for doubly insulated ground-less designs at higher voltages, but I'm not going to pretend that its not fun to think about). It gets its own test bench setup, since at those voltages you can't test on ESD benches ;)

hipot.jpg
That sounds like a great deal, looks like a lot of assembly, anytime wires or crimping is involved it's usually not quick.

I was just joking about the load but that's a very generous offer, I'm not sure I should be playing with those kind of voltages though lol. Maybe one day down the road I'll have you blow up a board for me so I might get the theatus certificate of safety. :)

Oh and I forgot to say earlier, if you are blowing things up make sure you hit record first so we can see, does sound like fun. :)
 
OP
theatrus

theatrus

Valuable Member
View Badges
Joined
Mar 26, 2016
Messages
1,319
Reaction score
2,551
Location
Sacramento, CA area
That sounds like a great deal, looks like a lot of assembly, anytime wires or crimping is involved it's usually not quick.

I was just joking about the load but that's a very generous offer, I'm not sure I should be playing with those kind of voltages though lol. Maybe one day down the road I'll have you blow up a board for me so I might get the theatus certificate of safety. :)

Oh and I forgot to say earlier, if you are blowing things up make sure you hit record first so we can see, does sound like fun. :)
One thing I've learned is if you have the wrong tools, you'll hate it. The LEDBrick build used Molex PicoLock connectors, which meant I had 8*2*2*4 = 128 crimps for just the LED harnesses alone, and I was doing it with what effectively was a fancy pair of pliers. Horrible idea, though the connectors do remain working to this day.

Use the right tool and its a no brainer. The biggest time is still the wire cutting and stripping.

The downside? For most connector systems, each terminal's tool will easily run $300+ for a single hand tool. Not a thing you want to buy for a single project. I don't have anything fancy like pneumatic strippers or such - cable production is the first thing that gets outsourced since its tooling intensive :)
 

Have you ever TRULY had a reason to panic in this hobby?

  • Yes

    Votes: 270 72.4%
  • No

    Votes: 95 25.5%
  • Don't know

    Votes: 8 2.1%

Online statistics

Members online
2,017
Guests online
4,865
Total visitors
6,882
Growcorals.com
Top