HMv4.0 to HMv4.1?


 
Complied with all suggestions and added the 100k resistors and caps. Definite improvement in that before the noise monitors jumped between red and yellow and oscillations were up to 2 degrees. Now the noise icons are just steady yellow and the most temp jump is .7 degrees. I ordered the parts for updated blower so as to try voltage mode to see if that helps further. However all temps are reading 10 degrees low now and wonder if that should be compensated by a different resistor value or an offset on configuration page? I am using the 22k pullup resistors. Thanks again for the help.
 
I spoke too soon. While it does read room temp about 10 degrees low at 60 degrees, it only reads 70 degrees in boiling water! Probes seem to be non responsive with the cap and 100k resistor. When plugged into heatermeter bypassing the mod the probes work fine. Any ideas?
 
Sounds like some resistance is somehow in parallel and that's giving you sort of this combination of the voltages which means you're just getting plain wrong readings. I'm not sure what your circuit looks like (schematic-wise) though so I can't say where the problem lies.
 
Bryan Mayland said:
Sounds like some resistance is somehow in parallel and that's giving you sort of this combination of the voltages which means you're just getting plain wrong readings. I'm not sure what your circuit looks like (schematic-wise) though so I can't say where the problem lies.
Click to expand...

I have the resistor and cap wired just like in Ralph's picture on previous page. Guess I'll just have to wait till a TC amp is developed and use my 4.2 remotely. The 4.0 works great and even without remote temp gauges just running the servo and fan through a common cable with 4.2 makes it worth all the effort I've put into it! Remote setup is only an advantage during the winter months up here. Thanks for the response.
 
Well, damn.. I read your post and checked my unit and have the same issue... I guess I jumped the gun on it being "easy" to add those filters, 'cause looking more closely at the schematic it seems the pullup resistor has to be on the probe side of the RC filter resistor, the way we have it the the pullup is on the ATMega side of that resistor.... I have to assume that is the problem....

For now you could just remove that resistor on the CAT5 probes, not sure if you will need to remove the cap too, you'd have the test (compare how the probe reads through the CAT5 jack to plugged directly into the HM). I will dig into it on my v4.0 board to see how difficult it will be to get the RC filter right and report back when I have progress... for now I am gonna remove the part of my post about the RC filters...
 
I got a chance to look at my HMv4.0 to investigate what went wrong with adding the RC filters. As I noted earlier, the RC filter resistor must be placed between the ATMega and the pullup resistor, unfortunately not in the convenient position at the end of the CAT5 wire. In order to accomplish this there is no other way than to cut the short trace between the pullup resistor(s) and the ATMega as shown in this picture:
HMv4.0Cut.jpg

(The probe inputs to the ATMega are the first four pins from the top corner near the probe jacks, Pit Probe being on the corner with food probes counting in from that. I marked cuts for the Pit Probe and first two food probes, leaving the last probe/ambient temp alone)

It's not too tough to cut the trace(s), even after the HM is built up. After you cut the traces those pins on the ATMega should be isolated from the probe jacks, do yourself a favor and check for lack of continuity before you proceed... I made the three horizontal cuts and installed the caps and resistors and thought I was good to go, the two food probes wire perfect but the pit probe was wonky... Took me a minute to realize the pit probe has the trace routed a bit different from the other probes and so you have to cut an additional trace, which is indicated with the vertical line. This trace is between the ATMega and the probe jacks, kinda tight but there's nothing else down there but the ground plane so just get a look at the trace with a magnifier and then cut roughly in that area with a razor blade/knife. Now you need to add a jumper wire from the Pit Probe to the pullup resistor since you had to cut that trace. I chose to jumper from the probe header instead of the pit probe jack because it was closer. Once that trace was cut and jumper installed the all probes were working nicely with RC filters in place. I tested them in hot tap water and compared to my HMv4.2 and readings were spot on.

Here is a picture of the resistors and capacitors installed on the board after I cut those traces.
HMv4.0RC.jpg


With all those resistors clustered in there at the ATMega it can get a little messy, a cleaner alternative would be to carefully solder in SMD 100K resistors between the ATMega pins and the pullup resistors. Unfortunately I didn't have any 100K SMD on hand so I soldered in a 10K for the pit probe just as an example.
HMv4.0RCSMD.jpg


With the traces cut, resistors installed and caps across the probe jacks as shown the RC filters are in place for both the onboard probe jacks and the CAT5 probe jacks as well.. and are working well...
 
Last edited:
I took a look at the picture of my HMv4.1 board and it seems adding the RC filters would be considerably easier on this one...
The pullup resistors are offset a bit and the traces between them and the ATMega are in the clear and would be easier to cut (and the pit probe only has one trace like the food probes on this board). Here is a pic showing the traces that need to be cut and drawings of where you would place the (through hole) resistors. In addition to this you would need to add 0.1uF capacitors across the probe jacks (from signal to ground).
HMv4.1RCCUT.jpg
 
Couldn't wait for SMD resistors so made the mod with TTH. My heatermeter 4.0 is in the house warm and dry while a rack of ribs is on the BGE! It works great! Strange thing is that when I tested it for about an hour I had no noise icons but when I later set it up on the BGE I am getting a yellow noise icon but pit temp is as stable as ever, no problem. I will do the new blower mod when parts come in and rewire using cat6 from house to deck and call it quits. Thanks for the excellent instructions and pictures, they made the mod easy.

2014-12-11Ribs_zpsb3a21f00.png
[/URL][/IMG]
 
I had a similar issue when i was making the aux. Tc board. I had no noise icons and when i attached it in the cold weather outside, i had noise, but stable temps.
 
John Bostwick said:
I had a similar issue when i was making the aux. Tc board. I had no noise icons and when i attached it in the cold weather outside, i had noise, but stable temps.
Click to expand...

I actually tested it in the cold! I had the blower and probes out laying on the deck, I think it was high 20's at the time ( I live near you in St. Charles ). No noise icons at all, then later using same cabling restarted it for the cook and got the icons. I must be right on the line but as long as it works I'm a happy guy!
 
St. Charles nice area, I work there.

I should have said that, I also did not get any noise at all while testing(hot or cold), but once I took it outside and started to smoke, I received the noise icon at Noise 5. It would appear and disappear and like you the temps were still good. hopefully the new boards I receive from OSH work. I will receive 3 with just the TC circuit(3.3v or 5v) and 3 with a TC circuit, 2 probe jacks, blower and servo connections, all on one board.
 
I have been watching the development of a TC board. Wish I knew enough to help, but I don't! I think it would be a welcome addition to the heatermeter. Good luck with the new boards.
 
John Bostwick said:
I had a similar issue when i was making the aux. Tc board. I had no noise icons and when i attached it in the cold weather outside, i had noise, but stable temps.
Click to expand...

Well, I have been testing things trying to get rid of yellow noise icons using my "high end" power supply when I plugged in an old Linksys wall wort for convenience. Sure enough no noise with wall wort whatsoever! I can't believe it, I have shielded cat 6 on the way that looks like I won't need. So, with Ralph's noise suppression mod you can run a 4.0 remotely and if you are getting noise try a cheap power supply!
 
That's funny. The power supply I have that is the lowest noise is a Netgear router (not router like we say today, I mean an old 10/100 hub) 12V supply.

I've also added the hm-noise firmware to the avr online firmware list if you want to see exactly the noise on each line, just flash that. To turn the graph off just flash back the normal snapshot.
 
Nice! But I'm not sure what I'm looking at! Here is a screen shot of what I get when I click on the 1 on the right side.

Noise_zps9eee5549.png
[/URL][/IMG]
 
The pit probe is 5, then 4 is food 1, 3 is food 2, 2 is food 3. 1 isn't used on a 4.0, 0 is the buttons.

Ideally you want the graph to be one number or bouncing between two consecutive numbers indicating <1 LSB of noise. HeaterMeter shows yellow noise icons for ranges above 5 LSB, then red for ... something way high. 20 maybe? You can check different power supplies and try to find one that gives you <1 LSB of noise which gives you the most accurate readings.
 
Here is what I'm getting with it correctly set up. This is with 2 patch cords and 30 ft of cat5. Looks like <1 LSB!

Noise2_zps12f6897c.png
[/URL][/IMG]
 
Heck yeah that looks good! Just as a general piece of information, the scale on the bottom is milliseconds. You can see in your original graph a dip every ~16ms = 60Hz = electricity mains frequency, probably some sort of bleed over on the unused ADC channel.
 
You must log in or register to reply here.

 

Back
Top