Kettle recommendations / is PT100 to 1-wire feasible?

Hardware
#1

I’m moving to a new apartment soon where boiling on the stove will no longer be possible so I’m looking to switch to an electric kettle with one of the new steam condenser accessories that people seem to be having some success with. (ie. here).

I’m considering various options: ranging from (1) a custom kettle from Spike brewing, with extra 1.5" TC ports for heating element and steam condenser, to (2) a Blichmaan Boilermaker w/ their element and an extra TC port for the condenser, to (3) an SS Brewtech eKettle, with an extra TC port for the condenser. Of the lot of these, the SS eKettle is actually surprisingly reasonably priced for coming with TC ports, a heating element and an integrated temp-probe (which is a PT100, more on that below)–probably because it doesn’t come with other fancy accessories like a sight glass and dial thermometer that I don’t really need…

Some questions:

  • Is it possible to get some sort of PT100 to 1-wire micro-controller so the spark can use the integrated temp-probe in the e-kettle? I like the idea of an integrated temp probe.

  • The heating element in the SS eKettle is advertised as being designed for solid state voltage regulation (SSVR) rather than SSR control. I’m guessing that it’s probably not an issue to use SSR though, right?

  • If I do go with one of the pre-built options and need to add an extra TC port for the condenser, any advice on this? Better to order a custom kettle and get the sanitary welds from the factory or just drill a hole and use a through wall fitting . This would be above the water-line if that makes a difference.

  • I was thinking of just using the Blichmann relay module, as an off-the-shelf solution rather than having to figure out a housing for my own module. Has anyone done this before? It looks to just be a standard 40A ~24-280 VAC / 4-32 VDC SSR with a housing and heatsink, so I can’t imagine why this wouldn’t work.

  • Any other electric kettle advice? I generally brew 20-22L batches, so I don’t need anything huge, but it would be nice to get something that could be integrated into a future HERMs system.

Thanks!
Austin

#2

I’m guessing a the first half of the PT100 issue could be solved with something using this IC (like this board, for example).

But what would then be the best way to send the readings over 1-wire to the spark (ideally emulating the DS18B20 reading format so I can just use it with the existing software)? Could this be done with an arduino nano? Or is there a more eloquent way to do it?

#3

RTD

Adding support for a onewire to pt100 interface in our firmware is the best way and something I want to do. When I can squeeze it in on my todo list is the biggest issue. All pro breweries use PT100, so I want to add support anyway.

The IC from maxim that you linked is the way to do it.

The IC you linked doesn’t use OneWire to communicate and therefore it is not easy to interface to the Spark.
Maxim has a 1-Wire thermocouple IC, but no IC for RTD sensors.

I have just made a new prototype a new very accurate sensor interface board for pressure sensors and RTD sensors, but it is designed to be extremely accurate and will be too expensive for this purpose.
It talks to the Spark over RS-485 instead of OneWire. It has its own microcontroller and two expensive 24-bit ADCs. Not the right solution here, you don’t need 0.0001C accuracy.

A different board with a microcontroller and cheaper ADC would be needed.
Emulating a OneWire slave with an Arduino is not allowed by Maxim, so it is not something I could sell. RS-485 would be better to use for communication in this case.

Kettle

If you DIY a kettle, we have all the parts and you probably end up spending less.
The only thing you need to control a heating element with the Spark is a solid state relay (SSR). I would stick with solid state, because you’ll use fast PWM want to switch on a zero crossing.

To build a custom kettle, you could use one of our round heating elements:
https://store.brewpi.com/mashing/stainless-steel-heating-elements?cat=37
You just need to make a hole with a 47mm hole punch.

We have these temp sensors to easily mount in a kettle wall:
https://store.brewpi.com/temperature-control/temperature-sensors/wall-mounted-onewire-temperature-sensor-for-kettle-or-fementor-bsp-231
https://store.brewpi.com/temperature-control/temperature-sensors/wall-mounted-onewire-temperature-sensor-for-kettle-or-fementor-npt-232

If you buy a kettle with integrated thermowell, you can slide in one of the simple 6mm probe OneWire sensors.

The blichmann module seems to be just an SSR in a housing indeed.
An SSR and heat sink in our shop costs 19 euro. The blichmann module 99 USD, whether that’s worth it is up to you.

#4

Thanks for the quick response. I didn’t realize that OneWire was proprietary and owned by Maxim. I guess that’s why I can’t find many RTD-1W solutions. A RTD to RS-485 breakout board would be awesome, especially if it also could control a relay or two. That way I could have a single wire run from spark towards a kettle control box.

I’ve used parts from your store before and have always been very happy with their quality. I will probably go the DIY route when I shift up to a fulls HERMs setup, but I’m a little time constrained at the moment, so am leaning towards a mostly off the shelf solution that require minimal modification on my part. Additionally, my current kettle is a bit under-sized and has BPT fittings, which don’t fit any of the parts I can get around me (I live in US but bought most of my equipment while living in China). Since I need to get a new kettle anyway, the SS eKettle that already has the element, valve, and TC clamps/blanks included suddenly becomes a surprisingly affordable option. I suspect they are making most of their money on the accompanying controller unit (which is a whopping $600).

What’s most attractive about the Blichmann unit is actually not the SSR itself, but that it comes with an enclosure, heatsink, and tails with L6-30 connectors. Especially given my lack of time, I don’t think I can beat that price putting something together myself. If you are able to produce a RTD ADC break-out board for the spark system at some point in the future, it would also be a super clean setup to just mount that in this enclosure with one wire heading back to the spark and an extra wire heading out to the PT100.

I will probably place an order for a 6mm probe and some other miscellaneous parts in the meantime (and leaving the PT100 board for a future upgrade once it’s out). I really like the TC probe, but it seems that it is 1" TC, whereas most of the kettles I’m looking at have 1.5" TC fittings.

Thanks again!

#5

1.5" and 1" TC both use 50.5mm OD I think.

If your kettle has 1/2" BSP ports, I have some custom triclamp adapters with integrated seal (no teflon tape required) for them that I still have to add to the store.

#6

Unfortunately, I think everything is 3/8" BSP, and honestly it might not even be standard BSP. China tends to use BSP for the domestic market, but sometimes I’ve found a lot of variation from one manufacturer to another–much lower QC than is used for export market. The adapter sounds like a cool part though!

I think you are right about the TC OD, so I probably will go that route when it comes time to make my order. Thanks for pointing that out!

I guess I have some other questions for if find myself with more free time before you are able to bring an RTD ADC board to market and want to experiment with hacking an interim fix together.

If I did use that adafruit MAX31865 board and an arduino nano for its ACD and RS-485 comms, this seems like it would be much cheaper than commercial RTD ADCs that support RS-485. Can the spark provide 5v power and RS-485 simultaneously over the RJ15 ports? I seem to remember it providing 5v/12v power for the pump control expansion boards, but I’m not sure if it could do the same with RS-485 protocol vs. OneWire.

#7

The RJ12 connector has 2 pins for RS-485, 1 for OneWire and 3 for power.

RS-485 communication has not been implemented on the Spark yet, because I didn’t have expansion boards yet that use it. But the new high precision board needs it so it will be added.

The Arduino (like every other processor) has RS232, but needs an interface IC to RS-485, a line driver basically.

#8

Ah gotcha. So I would either need an TTL to RS-485 breakout board or a dev board with built in RS-485 transceiver IC.

It seems like for two $10 breakout boards (ie. an RTD amplifier and RS-485 transceiver and an arduino lying around, I would have all the hardware I would need to experiment once you implement it in the firmware.

About how much time would you foresee between implementing RS-485 comms on the spark (and a protocol for requesting/sending PT100 temp readings) and having a PT100 breakout board in the store? Unless you foresee very little lag time between the two, this could be a fun project.

#9

While waiting for the serial comms, you could emulate a ds18b20. I use a esp8266 to receive a espnow msg from a custom ispindel which drives the emulated ds18b20 and talks to spark with onewire.

#10

@Elco resurrecting a long dead thread, but just wondering what the approximate timeline will be for for the breakout board with PT100 support. Guessing sourcing components hasn’t been fun over the last couple years, but I’m in no rush, just wondering whether to expect it this year, next year, etc.

Spark 4 looks pretty cool! Will probably upgrade once RTDs are supported.

I guess one question: will the RTD board also have any GPIO sockets that could be used to drive an SSR? I’d probably use the new Spark 4 purely for driving an electric kettle (and keep the spark 3 in charge of fermentation chambers), so it would be cool if I could use a single board for both the RTD and the SSR. Not the end of the world if that’s not in the plan, just curious.

#11

We recognized that firmware / hardware was a bottleneck for us, so I’ve spent the past month getting up to speed with firmware. This should leave Elco with more time for hardware development, and we’ll be re-evaluating long term planning accordingly.

Spark 4 GPIO is implemented using a stackable module, connected by generic pin headers. The Spark supports up to four attached modules.

We intend to use the same approach for other modules (eg. RTD), so you can mix and match.

#12

I would think pt-100 support would be way priority over designing boards for pressure sensing and volume levels. I know sooooooo many people who would make the switch and buy your products if u offered the support. Ds18b20 is janky and has very little quality tc coupling solutions I honestly can’t stand how slow they are to react.

#13

I’ve actually always had pretty positive experiences with their responsiveness to questions, the timeliness of pushing fixes/adding features in software, etc… Designing hardware to be produced at scale always takes more time than tweaking software, and the last couple years have had some exceptional circumstances re. supply chains for electronics. For me RTD support is more useful, but it’s not hard for me to imagine other customers for whom volume sensing, etc. is just as important.

To clarify, my question was whether the RTD board will be similar to the onewire board–which looks like it has 4 onewire ports on one side, then on the other side some of the green gpio ports (which I would use for controlling relays). I’m wondering whether the RTD board will have some number of RTD ports on one side, and then once again some of the green ports for controlling relays. This would be beneficial for me because I could use a single expansion board, and then fit the setup in a shallower project box (vs. if I needed to stack 2 expansion boards to get RTD and relay ports). That said, I can see how for larger systems it may be disadvantageous to have more multipurpose and less specialized expansion boards.

#14

Board design is not yet at a point where we can give a reliable yes/no as to whether we’ll add relay ports to the RTD board. It makes sense to have a RTD + IO counterpart to the OneWire + IO module, but only if it doesn’t come with a significant cost / complexity increase.

1 Like
#15

I plan to make the RTD board equivalent to the OneWire board, with the same output driver, just different inputs. Another issue currently is actually buying ADCs, everything is out of stock.

#16

That sounds perfect for my needs. I imagined that IC supply chains were probably a limiting factor. Hopefully will improve soon, but the Shanghai lockdown can’t be helping. At least Guangzhou and Shenzhen still seem to be open. Good luck!