Auber Pid odd behavior in Auto-Mode 100% or 0% and nothing else

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

biertourist

Well-Known Member
Joined
Oct 8, 2012
Messages
888
Reaction score
148
Location
Woodinville
Since I reset all of my Pid settings and performed a new autotune a while back, I've noticed that when I have my pid controller in Automatic mode that it's rather "Bursty"- the temp jumps when the element comes on and then it rapidly falls.

When I bump the A-M button, while it's still in Automatic mode (A-M light illuminated) I can see why this is happening - the PID controller is making the output either 100% on or 0% on and nothing in between for some reason.


When I first got the pid it would be a very nice slow ramp and wouldn't overshoot hardly ever, it would vary the heat output and slow it down from 100% as it approached my set temperature.


Now when my Pid is in automatic mode it's output is either 100% or 0% and nothing in between...

Anyone got any ideas what could account for this?



Adam
 
I'm assuming you have a SYL-2352. When you're asking technical questions like this it kinda helps to provide all of the pertinent information, and as much detailed information as you can.

Now, continuing to assume it is a SYL-2352, the fuzzy logic can be fickle after autotune and the P, I, D, t values can have some strange relationships not typical of regular PID controllers. Reset to factory values and run autotune again.

Why did you perform a new autotune? If it was working well before, just change it back to your previous values.
 
I'm assuming you have a SYL-2352. When you're asking technical questions like this it kinda helps to provide all of the pertinent information, and as much detailed information as you can.

Now, continuing to assume it is a SYL-2352, the fuzzy logic can be fickle after autotune and the P, I, D, t values can have some strange relationships not typical of regular PID controllers. Reset to factory values and run autotune again.

Why did you perform a new autotune? If it was working well before, just change it back to your previous values.

Yes an SYL-2352.

I performed a new auto tune because I moved my temp probe from the bottom of my sight glass to the center of my kettle AND because I have a 3 way switch that physically turns off the power out to the kettle and I had been turning it off and really confusing the hell out of the PID's fuzzy logic algorithms.

Is there some magic way to reset all settings to factory default or do you manually have to go setting by setting and reset them?


The plot also thickens on this issue: even when I set the pid into Manual Mode now let's say at 60% heat output, the heat "pulses" / is "bursty". I set by boil kettle down to 60% after it came to the boil on Sunday and it boils violently for 2-3 seconds and then calms down for 2-3 seconds and then repeats over and over again, much like it's pulsing between 0% and 100% rather than keeping a constant 60%.


Adam
 
The cycle time on your PID is probably set to about 6 to 10 seconds rather than the shortest possible cycle time. In manual mode at 60%, a cycle time of 10 seconds will give you about 6 seconds at full power and 4 seconds off, a little shorter for shorter cycle times (e.g. 3s on, 2s off at 5s cycle time). In auto mode, it will vary the amount of the cycle time that it's at full power for based on the PID algorithm, which could easily be several seconds on before several seconds off as well.

These PID controllers can only switch the output on or off, they can't set anything in between. But with the cycle time set to the shortest available time, they can switch on and off a couple of times a second, which does a reasonable job of mimicking intermediate power outputs with the kind of thermal mass we have in our systems. But at cycle times greater than a few seconds (on and off once every <n> seconds) the thermal mass of the system isn't enough to damp out the pulses of power, and so you see the on-off boiling.
 
Yes it sounds like cycle time t is probably longer than it needs to be. The problem is that it can sometimes throw the fuzzy logic off when you change it back to the minimum of 2 (actually if you set P to 1 you can set the minimum even shorter to avoid the pulsing).

One other suggestion is to contact Auber about technical support. They are usually very responsive and helpful. That's the main benefit in getting an Auber controller instead of the rock bottom China ones written in Engrish.
 
Yes it sounds like cycle time t is probably longer than it needs to be. The problem is that it can sometimes throw the fuzzy logic off when you change it back to the minimum of 2 (actually if you set P to 1 you can set the minimum even shorter to avoid the pulsing).

One other suggestion is to contact Auber about technical support. They are usually very responsive and helpful. That's the main benefit in getting an Auber controller instead of the rock bottom China ones written in Engrish.
The auber units are made in china too ;) as you say you are paying extra $25-30 for the directions and support from auber. otherwise you can buy the exact same pid labeled as the XMT612 direct from china for $30 shipped...

its possible that there are actual engineers at auber that designed this unt and china cloned them after auber sent the plans there to have them made cheaper... however we are really just guessing. the fact that most of the other products aubrins sells are just generic product from china they relabel as thier own suggest otherwise at this point.
 
The auber units are made in china too ;) as you say you are paying extra $25-30 for the directions and support from auber. otherwise you can buy the exact same pid labeled as the XMT612 direct from china for $30 shipped...

its possible that there are actual engineers at auber that designed this unt and china cloned them after auber sent the plans there to have them made cheaper... however we are really just guessing. the fact that most of the other products aubrins sells are just generic product from china they relabel as thier own suggest otherwise at this point.

I don't think that's really appropriate for this thread. Why are you so stridently against Auber? I've never heard of anyone having a bad experience with the company.
 
I don't think that's really appropriate for this thread. Why are you so stridently against Auber? I've never heard of anyone having a bad experience with the company.

I only stated the facts You actually started on THIS topic with the post I commented on NO?...I just added the rest of the picture you painted with your derogatory china engrish comments suggesting the auber products were something different.

Since you asked. I dont care for them because there is a false persona here that their products are better at 2 or 3 times the normal prices you would find the exact same product made by the exact same chinese manufactuers elsewhere... People need to know the truth that they are just a distributor who doubles the prices of the same stuff you can find anywhere else for much less.

I didnt say they are a bad company... Only that they buy most of their stuff in bulk from china and mark it up and sell it for about double as sold direct.. I have in other threads shown this many times but many are still unaware and spend much more thinking its a different product they offer. they are often just a reseller/middleman.

I for one wouldnt want to buy a chevy spark only to get it home and find it was in fact a relabeled deawoo made in korea and marked up 25% by GM before they passed it off to me implying it was a chevy product made by chevy... To each his own I guess...I dont think GM is a bad company for this but I just assume buy something else rather than stick my head in the sand...

Id rather not derail this thread any further so let please let it rest...
 
Maybe now in 2015 you can get similar hardware direct from the factories but the software, support, and convenience all add value to the product, which is still very reasonably priced. I've only had good experiences with Auber parts and support and recommend getting in touch with them for any issues.
 
Maybe now in 2015 you can get similar hardware direct from the factories but the software, support, and convenience all add value to the product, which is still very reasonably priced. I've only had good experiences with Auber parts and support and recommend getting in touch with them for any issues.

not similiar products... The very same products, Their probes, switches, indicators, contactors ,ssrs enclosures , heat sinks timers, and some of their pids are all available direct from the real manufacturers or other distributors for much less. usually half the price or less than they mark things up to.. For example an Aubrins ssr is really a magar brand ssr available for $9 shipped on ebay new but they are $15 plus another $5 shipping from auber making them double the price...thier generic $10= shipping heat sinks are $3 with free shipping elsewhere..s its great if you need someone to hold your hand on how to use something but it adds up quick so I just like people to know the facts about the products before they spend $500 there only to find they could have got all the exact same stuff for closer to $200 if they just wait an extra week or two for shipping times. auber buys this stuff in bulk from china for pennies on the dollar and then charges double the retail while duping many into thinking their product is superior quality... Ironically Auberins has dropped their prices a bit in the last year on the smaller stuff... hmmm yeah the support is better from auberins..I wont argue that... I just want the facts to be on the table about the products... much of what they offer is the same generic stuff just repackaged by auberins.
 
The cycle time on your PID is probably set to about 6 to 10 seconds rather than the shortest possible cycle time. In manual mode at 60%, a cycle time of 10 seconds will give you about 6 seconds at full power and 4 seconds off, a little shorter for shorter cycle times (e.g. 3s on, 2s off at 5s cycle time). In auto mode, it will vary the amount of the cycle time that it's at full power for based on the PID algorithm, which could easily be several seconds on before several seconds off as well.

These PID controllers can only switch the output on or off, they can't set anything in between. But with the cycle time set to the shortest available time, they can switch on and off a couple of times a second, which does a reasonable job of mimicking intermediate power outputs with the kind of thermal mass we have in our systems. But at cycle times greater than a few seconds (on and off once every <n> seconds) the thermal mass of the system isn't enough to damp out the pulses of power, and so you see the on-off boiling.

THIS! This is what I've been waiting for someone to say; this matches my behavior exactly!


Adam
 
Yes it sounds like cycle time t is probably longer than it needs to be. The problem is that it can sometimes throw the fuzzy logic off when you change it back to the minimum of 2 (actually if you set P to 1 you can set the minimum even shorter to avoid the pulsing).

One other suggestion is to contact Auber about technical support. They are usually very responsive and helpful. That's the main benefit in getting an Auber controller instead of the rock bottom China ones written in Engrish.

So the setting that I'm looking for is just "t"?

[edit]checked the manual; yep, it's just "t". "2" is the lowest cycle time, I'll check tonight.

Adam
 
It can be tricky, sometimes even when t = 2 you get a longer cycle time than 2 seconds when in manual mode.

In order for me to get rid of any noticeable pulsing during manual mode boil I had to set P = 1, which only then allowed me to set t = 1 and actually work at that short cycle time. I do have a dedicated controller for the Boil Kettle so I don't need any precise PID algorithms. I'm sure you can tune it to run fine without the "fuzzy" logic, though.
 
It can be tricky, sometimes even when t = 2 you get a longer cycle time than 2 seconds when in manual mode.

In order for me to get rid of any noticeable pulsing during manual mode boil I had to set P = 1, which only then allowed me to set t = 1 and actually work at that short cycle time. I do have a dedicated controller for the Boil Kettle so I don't need any precise PID algorithms. I'm sure you can tune it to run fine without the "fuzzy" logic, though.

I do NOT have a dedicated PID for boil control, although I plan to move the boil to dial / rheostat control in the next 6 months so switching to P=1 and back again right now is just too painful.


I think this WAS my problem: my "t" setting was set to "10" which pretty much tracks with exactly what I've been observing. I won't have the chance to test it out at least until the weekend but, I'm pretty happy to have found this setting.

Thanks everyone for your feedback; HBT saves the day again. And to jump on AugieDoggie's comments, maybe if HBT keeps providing the solutions to my PID problems the extra $$ I spend on Auber pids isn't worth it.

I know for some people the peace of mind of having support is worth it, but I think I'd rather save the $$ and figure it out with the assistance of free online forums filled with smart people.


Adam
 

Latest posts

Back
Top