www.brultech.com

Hello All,

I wanted to confirm if the GEM uses an actual 1 wire chip to perform its temperature functions. Opposed to using an emulated software solution to do the same.

It's a software implementation and it doesn't have a strong pull-up so parasitic may not work. For parasitic (no power wire), you can try adding a strong pull-up between 5V and Data.

I've just moved a 1-wire bus over from a WEL server to the GEM. I know all the sensors are/were working correctly before I moved the bus over. They are all parasitic and I'm wondering what needs to be done (if possible) to get them working.

All the sensors were discovered but they are all reading 185.0 F.

You mention doing a strong pull-up between the 5V and Data pins. I'm not exactly sure what this means or how to do it. I'm willing to try anything if you can give me instructions on what needs to happen.

slords wrote:I've just moved a 1-wire bus over from a WEL server to the GEM. I know all the sensors are/were working correctly before I moved the bus over. They are all parasitic and I'm wondering what needs to be done (if possible) to get them working.

You mention doing a strong pull-up between the 5V and Data pins. I'm not exactly sure what this means or how to do it. I'm willing to try anything if you can give me instructions on what needs to happen.

Whoaaaa this is a really old thread!

Your first task is to wire one single 1 wire sensor to the GEM as outlined in the full installation manual. Next, to enroll and detect each sensor (one at a time) once done you may deploy to where ever you need them.

I've got all the sensors enrolled already. They are all detected fine. The issue is whether I connect them one at a time or via the bus they only read 185.0 F. The sensors only have 2 wires and are parasitic. They have been working fine for years hooked to a different 1-wire system. If I hook them back up to that system they read the correct values.

ben wrote:It's a software implementation and it doesn't have a strong pull-up so parasitic may not work. For parasitic (no power wire), you can try adding a strong pull-up between 5V and Data.

I'm wondering what this means and if I can do this in order to get the sensors working.

Once all of the 1 wire sensors have been properly enrolled its up to you to decide what wiring topology to use. There are pros and cons of using a star, stub, and linear which is dictated by various elements like length and type of wiring.

Regardless, all of the sensors are going to be connected to one central point somewhere.

As I've already said. I already had a 1-wire bus installation that has been working great for years. I just moved the bus over to the GEM.

It is a straight 80 ft run with 8 sensors installed. All the sensors are already registered to the GEM. I'm just not getting valid temperature data out. They all ready 185.0 F (185!). That number is constant and unchanging. If I remove one sensor and connect it directly to GEM I get the same value. The actual temperature should be between 60-80 F.

slords wrote:As I've already said. I already had a 1-wire bus installation that has been working great for years. I just moved the bus over to the GEM.

It is a straight 80 ft run with 8 sensors installed. All the sensors are already registered to the GEM. I'm just not getting valid temperature data out. They all ready 185.0 F (185!). That number is constant and unchanging. If I remove one sensor and connect it directly to GEM I get the same value. The actual temperature should be between 60-80 F.

My earlier reply to you was to address your question about how it should be wired up. That was to first decide on the type of topology that best suits your environment. Regardless of that, the GEM is a three wire system and requires ground, power, and data.

You only have two wires which I have never seen in a true Maxim / Dallas 1 Wire system. The GEM only supports 3 wire DS18B20 sensors and the other which I don't recall off the top of my head.

If I have to replace the sensors with the 3-wire version I will. But Ben states that if the sensors are parasitic that adding a strong pullup between 5V and Data might fix the issue. I don't know what this means or how to do it. I'm willing to give it a try but I need directions on what/how to do it.

If I try and it doesn't work then I'll end up replacing the sensors.

slords wrote:If I have to replace the sensors with the 3-wire version I will. But Ben states that if the sensors are parasitic that adding a strong pullup between 5V and Data might fix the issue. I don't know what this means or how to do it. I'm willing to give it a try but I need directions on what/how to do it.

If I try and it doesn't work then I'll end up replacing the sensors.

I believe Ben is indicating to place a resistor of defined value at the end of the sensor which connects both 5 VDC & Data. Going off rough memory that resistor was 10K plus but you can search the forums for the exact value.