PID vs. Barksdale 0.05bar deadband pstat > now going dual pstat config

[miKe mcKoffee]

Having a heck of a time making up my mind, other than knowing I want to get rid of the Mater pstat. If going PID would not be used to change temp/boiler pressure on a regular basis more like pstat set it and forget it. And internally mounting not a problem with the inside lower back area only ~89f.

On the other hand talked to Tomas of Gimme Coffee today and the Barksdale pstat sounds very interesting. They custom ordered them and have been using them a couple years flawlessly. Supposedly very quiet with 0.8 to 1.6bar adj range and 0.05bar deadband, now that's tight. Though rated 15A he highly suggests driving the heater via SSR, which isn't a problem of course plus Bricoletta already has a Siemens SSR.

For some reason I'm not sure a PID would hold the boiler as temp stable under heavy use as a 0.05bar pstat. Am I wrong?

[erics]

Well Mike, strictly by the numbers, 0.05 bar is 0.725 psi which equates to about 1.3 - 1.5 degrees F according to my Steam Tables Handbook.

Bar Degrees F


0.70 239.7
0.75 241.3
0.80 242.9
0.85 244.4
0.90 245.9
0.95 247.4
1.00 248.8
1.05 250.2
1.10 251.6
1.15 252.9
1.20 254.2
1.25 255.5
1.30 256.8
1.35 258.1
1.40 259.3
1.45 260.5
1.50 261.7


I would say that a PID may do better because surely the pressurestat is an on-off device but potentially lots of parts changeouts and adaptors for how much gain? What does the Siemens SSR in the Bric do now - pulse the heating element? Mr. Barksdale is knocking (barking) at your door and me thinks you should open.

Eric S.

[cannonfodder]

Do you have a heat exchanger or double boiler? I can't remember but I want to say HX.

[miKe mcKoffee]

Do you have a heat exchanger or double boiler? I can't remember but I want to say HX.

Rotary Bricoletta HX, 1.5L boiler, 1800 or 1900W heater depending on production run which is unknown. (According to 1st-Line the heater element has varied production run to production run)

I would say that a PID may do better because surely the pressurestat is an on-off device but potentially lots of parts changeouts and adaptors for how much gain? What does the Siemens SSR in the Bric do now - pulse the heating element? Mr. Barksdale is knocking (barking) at your door and me thinks you should open.

Eric S.

It's because a PID is not a strictly on-off device that I'm not so sure it would hold the boiler temp as constant under use. Idle I'd expect a PID to maintain a tighter constant boiler temp. But as soon as leaching heat from the boiler either pulling a shot and/or steaming (which I do simultaneousely when making caps though just shot more common use) a pstat will give full heater power trying to maintain boiler temp vs. PID pulsing power trying to avoid overshoot until "it thinks" it needs full power to keep up. (Or so my limited understanding of PIDs goes) From using PID with Silvia temp would drop a fair amount before PID gave full on power but don't know if PID can be tuned for instant full power to operate more like a very tight deadband pstat.

The Bricoletta pstat drives SSR which drives heater, same as PID would drive SSR.

With the option of 0.05bar better than 1C deadband pstat the case for PID is it more reliable than a pstat when pstat not under the stress of directly driving the heating element? From the standpoint of if fails "on" causing constant heating and hence potential flooding from purged steam condensation. The Bric' has a thermal high limit trip but don't really know if it will help prevent flooding or just designed to prevent heater burn-out from lack of water which wouldn't be the case with direct connect with heater constant on. I could rig the heater full on and test it but am very hesitant fearing "what if" I fry something.

While I believe a PID may be more reliable from a failure standpoint, it's not an option unless it will perform as well giving full heater power when temp falls below set point during production use.

I hear ya Eric, listen to the Barking and just do it.

[cannonfodder]

PID on a HX machine may have a relatively low return on investment. I am not sure how much of an improvement you would get since the HX and group have a much bigger impact on the temperature than a tight deadband PID. On a double boiler, I would PID the brew group and Pstat the steam group personally. If you have a blown pstat and the PID vs. pstat price is relatively close, yes I would probably go with a PID. But to modify the machine with a perfectly good pstat, I would pass.

I had actually brought up the discussion at Barry's holiday party. The impression I got was don't bother. When steaming, the PID produces a jagged saw tooth profile as it switches on and off. When steaming, you want to heater to kick on and stay on for the session. The brew stability is essentially unchanged.

[miKe mcKoffee]

PID on a HX machine may have a relatively low return on investment. I am not sure how much of an improvement you would get since the HX and group have a much bigger impact on the temperature than a tight deadband PID. On a double boiler, I would PID the brew group and Pstat the steam group personally. If you have a blown pstat and the PID vs. pstat price is relatively close, yes I would probably go with a PID. But to modify the machine with a perfectly good pstat, I would pass.

I had actually brought up the discussion at Barry's holiday party. The impression I got was don't bother. When steaming, the PID produces a jagged saw tooth profile as it switches on and off. When steaming, you want to heater to kick on and stay on for the session. The brew stability is essentially unchanged.

Guess I should clarify my why. Direct shot temp control is not my reason for considering PID or tighter deadband PID. HX surf is and will be HX surf regardless. Though tighter deadband via tighter pstat or PID may improve intra-shot performance it's not the motivating factor. Primary reason considering PID is probable lower failure rate and hence less likely chance of failure heater stuck on causing continuous copious steam and ensuing condensation and flooding with direct connect. Second reason Mater actual setting tends to float a bit though it's a minor consideration factor. Recent CG thread on 24/7 operation vs. timer Itinerant proposed an interesting safety method for the potential flooding issue, namely using the pstat as a high pressure cutout inline with PID. I further wondered if two pstats could be used on a 'T' for the same purpose. First pstat high limiter wired inline with second main operational pstat.

I know the whole 24/7 operation issue has no clear cut consensus on what's better for the machine. I tend to give some credence to Terry's opinion that 24/7 operation is better for the machine and results in less scale build up. Especially the less scale build part since direct connect HX not that simple to descale. Since cost isn't the issue would prefer leaving on 24/7 but currently don't like the idea of a pstat failure potentially causing flood. (Did the math and comes to about $3 per month on 24/7 with the Bric' idle, ceramic fiber boiler insulation including ends resulting in approx' 3% heater duty cycle)

Cannonfodder, other than possible lower failure rate why would you go PID over tight deadband pstat? Can a PID maintain boiler temp as effectively as a tight deadband pstat while pulling shots and steaming?

Whether PID or tight deadband pstat I really like the idea using existing Mater pstat as high pressure limiter.

[cannonfodder]

On a HX machine, the only real reason I would PID it would be if my pstat was burnt out and the PID to Pstat cost was the same. I don't think a PID adds any performance boost to a HX. Keep in mind this is hearsay and personal theoretical observation, I have never PID'd a machine, in part because I own two heat exchanger machines. Others would be better qualified to verify my personal observation.

[greedus]

Go with the Barksdale. I have a HX machine, Cimbali Jr. , and the Barksdale is working flawlessly. It is silent and hold the temp very well. It presents to be well built and and is a good fit for the machine given over all cost and instalation effort.

I am very happy with it. It has turned out to be a great solution for the Cimbali HX. Every so often I pull a shot with out a flush and to my palate it is not too bad, very exceptable. Mind you, I have the p-stat set to .09,

regards

pete

[miKe mcKoffee]

While I believe a PID could be programmed to act more like a pstat giving full heater power and PID could be safely mounted internally in the Bricoletta, I ordered a Barksdale 0.05 deadband pstat today. Since either way planned to use the existing Mater inline as a high pressure limit safety PID installation would have actually been easier. Future boiler pressure adjustability made the Barksdale a better choice. Specifically would need to remove both the top cup warmer and the body wrap to get to PID to adjust instead of just the cup warmer. Removing the body wrap requires dragging Bric' out from under the counter while just removing cup warmer does not. Externally mounting PID is not an option I chose to entertain with the Bric'.

It will be interesting rigging a boiler T for mounting two pstats. Plan is to have the Mater adjusted up around 1.4bar or so as a high limit safety wired inline before the Barksdale main operational pstat. Since the boiler pressure should never reach the Mater's off point during idle or operation won't be clicking on and off so that noise eliminated. Unknown is how always being in the on position will affect the Mater long term. I believe will have virtually no affect on the contacts, especially since low current. (heater driven by SSR which will now be triggered by the Barksdale pstat)

This way if the Barksdale pstat should fail in the on position the Mater would kick in well before over pressure condition venting steam with ensuing condensation and flooding. 24/7 operational concerns pretty much eliminated or at the least highly minimized.

[JonR10]

...I ordered a Barksdale 0.05 deadband pstat today.

If you don't mind me asking - where did you order that from?

[miKe mcKoffee]

If you don't mind me asking - where did you order that from?

Gimme Coffee 877-446-6325. Tomas is their espresso service guy if you want to talk to him about it. Otherwise to order ask for Janet, Barksdale pstat $42 plus shipping. Tomas told me he could only get them made with 1/8" fitting so I'll (you'll) likely need an adapter.

[JonR10]

Gimme Coffee 877-446-6325. Tomas is their espresso service guy if you want to talk to him about it. Otherwise to order ask for Janet, Barksdale pstat $42 plus shipping. Tomas told me he could only get them made with 1/8" fitting so I'll (you'll) likely need an adapter.

Thanks Mike!

[Java Man]

Jon, if you decide to install a Barksdale in your Wega (assuming you are still running it) could you please post your impressions? I'm considering the same.

Cheers,

Rick

[another_jim]

Guess I should clarify my why. Direct shot temp control is not my reason for considering PID or tighter deadband PID. ... Primary reason considering PID is probable lower failure rate and hence less likely chance of failure heater stuck on causing continuous copious steam and ensuing condensation and flooding with direct connect.
...
Whether PID or tight deadband pstat I really like the idea using existing Mater pstat as high pressure limiter.

The Mater is a complete joke as a safety; it's primary failure mode is for the membrane to get stuck on full heat, especially in the first hours after a long off period (when the membrane is frozen in that position). This is easily fixed without replacement, but the one in my Tea relies heavily on the old tea kettle safety. Overall, there is no doubt that Espresso machines are far happier running 24/7. All of my leaks, pstat failures, relay failures, etc, happen in the first hour after firing up returning from trips. The Bric is probably better assembled by now (after a few years of outraged howling by the German espresso contingent got the bugs out), but it still uses the same parts bin as all the other e61 boxes.

PID/SSR systems tend to be more failure prone in the first few months, as electronics, heat sinking, or wiring faults are exposed, but after that they are usually immortal. Pstats wear out over a few years of 24/7 operation -- the contacts, especially if they are carrying current, and the membrane. On most I've heard about (the Barksdale is new to me and sounds great), the deadband tends to double over the course of a few years.

[miKe mcKoffee]

Hmmm, I wondered whether rigging the Mater as a high limit safety would work in practice. Since it would always be in the on position, whether machine on or off, seems like the chance it would work as a safety if/when called upon to turn itself and the heater off would be chancy at best. So guess I won't bother with the dual pstat T installation idea. Sounds like what's really is needed as a safety could be 260f or so thermal cutout inline before the pstat. The Bric' actually has one already but I don't know its trigger temp. Whether designed to protect against over pressure and continuous steam venting or a higher temp to only protect heating element from lack of water. Talked to Jim @ 1st-Line and he wasn't sure either. Maybe I will just have to by-pass the pstat and test to find out.

[another_jim]

Hmmm, I wondered whether rigging the Mater as a high limit safety would work in practice. Since it would always be in the on position, whether machine on or off, seems like the chance it would work as a safety if/when called upon to turn itself and the heater off would be chancy at best. So guess I won't bother with the dual pstat T installation idea. Sounds like what's really is needed as a safety could be 260f or so thermal cutout inline before the pstat. The Bric' actually has one already but I don't know its trigger temp. Whether designed to protect against over pressure and continuous steam venting or a higher temp to only protect heating element from lack of water. Talked to Jim @ 1st-Line and he wasn't sure either. Maybe I will just have to by-pass the pstat and test to find out.

If the heat is stuck at on, the safety valve will pop and blow lots of steam. On a plumbed machine, that'll be it. On a pourover, the tank will eventually run dry after a few hours, and the fusible link will melt, breaking the circuit. If you don't intend on coming back in 24 hours, turn the machine off. The kink is definitely set higher than the safety valve, since the machine is doa until it's replaced. Some machine's have resettable pop thermostats (like the expobar), and these will go around the same point as the safety valve. This type of trip thermostat is a common espresso machine part (you have one on the Silvia probably) and they come at various temperatures -- you could check with EPNW or some other supplier for the temperature selection.

[cannonfodder]

Personally, I would put a thermal fuse on the machine. I have burned out one on my Isomac. They are easy to install and easy to replace. My Gaggia lever has a resettable thermal breaker which is even easier than a fuse.


red button in center is a resettable thermal breaker on my lever machine


Fried thermal fuse on my Isomac

[Ken Fox]

Regarding the initial question, PID vs. the Barksdale pstat, I have no opinion for any particular machine being as I don't have any idea how likely one is to reach better temperature stability with a boiler PID in HX machines other than those I own, with a PID vs. this pstat. I also have no experience with this pstat itself. Eons ago, before I PID'd my vibe machine (the rotary has been PID'd for more than 2 years), I discussed the then novel CEME (supposed) low deadband pstat with Michael Teahan, my usual referral expert on such matters, and he mentioned the Barksdale as an alternative, mentioning Gimme as the place to call. I never bothered, but for no good reason. Without doubt, a truly functional low deadband pstat is going to be a much easier and less time intensive installation than is a PID.

When the CEME proved to be worse than useless on the old vibe machine, I went directly to a PID, and the implementation on my old vibe machine has been, if one bases success on results obtained, better than on my newer rotary machine. I am able to preserve the HX hump and to get somewhat better intershot temperature stability on my 10+ year old vibe machine than I can get on my current vintage rotary Cimbali Junior which doesn't seem to have much interest in humped profiles whatever I do with it. But I digress.

There are several issues here that are all getting mixed up and the "right" decision about what to install for one person may be the wrong decision for another. I would not recommend mounting a PID inside of a machine exposed to the expectedly hot temperatures one would encounter near a boiler. Barry Jarrett has pointed out that there is an "internal reference junction" inside of PIDs, or at least inside of the Fuji PXR3, and if this is not ventilated to outside air the behavior of the PID controller can be erratic. I would defer explanations of this to Barry, however he convinced me that I could not expect stable performance with a controller that was not ventilated, who tried such installations and found them wanting. What this means is that you need to install the controller external to the machine. In order to do this well you need a project box, and there are a plethora of options out there of aluminum project boxes that are only a little bit larger than the size of the Fuji and similarly sized controllers. Mouser.com is a good source for stuff like this. There are also plastic box options if you don't mind looking at them.

The SSR can likely be mounted using the machine's chassis as a heatsink, which is what I've done in both of my installations. I have placed a TC next to the endplate of these SSRs, as recommended by an SSR mfr's tech support, and temperatures measured showed that the machine chassis worked well as a heatsink.

So, the installation can be done in a way that is not too ugly but one has to locate the project box in a place that with a water accident (e.g. such as a drip tray overfilling, and these things do happen and have happened to me) the controller will not be exposed to water, also the controller itself should be fused, which is easy enough and might add a dollar to the overall project cost. But only you can determine if having a project box outside of your machine will mar its appearance to a point where you can't live with it. I do not find either of my installations to be visually offensive but that is a personal call.

The PID, as Jim has pointed out already in this thread is apt to be very reliable if you don't have an early electronic component failure. The Mater and the CEME are both glorified pieces of sh*t, if one is talking of reliability, quality, and service life. I know nothing of the Barksdale and can't comment on it.

OK, that is one issue, basically the difficulty and the appearance of the end result, of putting in a PID vs. putting in this particular pstat.

Here's another issue; do you want to control your shot temps? Now, there are ways to control shot temps and these are going to be very machine dependant. I'm sure that Dan Kehn has his Elektra nailed and can get within a degree or two on his stock machine just by water dancing and the like. Even he would say that the first shot he gets is more of a crapshoot. I've demonstrated ad infinitum that with a PID on my machines and with a low temp boiler setting and with a small flush I can reduce the variability of the hard to control first shot. I have to say however that in extended testing (yes, I have "published" only a small fraction of the shot temperature graphs I've accumulated over time and have tried a whole lot of scenarios with varying the flush volumes, boiler temps, interval between flush and shot pulling, etc. etc. etc. but have not "published" the results) there is often an "outlier" shot that does not "behave." By this I mean that I might do a series of 10 "walk up" shots, and 9 of them look great and one of them is for sh*t, off by 2 degrees or so for no apparent reason. I can't explain such behavior as it does not appear to follow any pattern. In my own usage such a shot would have at least a 50-50 chance of going into a milk drink, where it would likely not be noticeable, and maybe the other ones are sink shots, I just don't know.

How any of this compares to what one would get on a Barksdale pstat, or how it would work on your machine I do not know.

OK, how about if you want to control your shot temps? Dan K. has the flushing ritual down and he probably doesn't care about this sort of stuff, but let's say you have decided that your SO Harrar Horse makes its best shots at an extraction temperature below 200F, and you have other blends you like that "prefer" 202F? With a pstat that means either getting your flushing ritual down pat, if you can, or it means adjusting the boiler temp. Do you want to have to take a side panel off your machine and get out your small screwdriver every time you need to adjust the boiler temp because that is how you adjust shot temperatures? I don't think so as that is an enormous PITA. Trust me, you won't do it. So, you will be stuck with flushing as your only way to adjust shot temps and it may well work for you.

Since the regimen that works best for me involves fairly low boiler temps, that might not be the best for frothing, what I tend to do is to leave the boiler set at a temp that is best for straight shots on whatever I'm pulling at the moment, and if I want to make a milk drink I bump the boiler temp up a few degrees which allows a good milk drink and attendant frothing to occur. Then in 4 seconds I reduce the temp back down if I think my next shot will be a straight one. Simple. This works well for me because a PID temperature controller is easy to reset temperatures on.

I've found that what I'm doing is easy for me to work with. I don't know if it will be easy for you. I have had absolutely no regrets about having PID'd my two machines.

YMMV.

As to issues regarding safety, I have not experienced an over temperature failure of my PID's. From what I have learned, however, at least on Cimbali Juniors, if you have a pop valve incident the effects are going to recur repeatedly every 10 or 15 minutes and each time water is going to be released and you may get a mess. After the boiler is empty the thermal safety device on the element will kick in. Older machines have an inline fuse of sorts that will ultimately "melt" and need replacement of the part. Newer machines have a pushbutton reset on the side of the outsides of the undertray electronics box (a red button). A PID if it fails is most likely going to be the result of an SSR that is "stuck on." If this happens you will have to rely on the boiler element safety device referred to earlier in this paragraph. Other failure modes are apt to result in the PID shutting off power to the element leaving the boiler water at a lukewarm temperature.

ken

[miKe mcKoffee]

I would not recommend mounting a PID inside of a machine exposed to the expectedly hot temperatures one would encounter near a boiler. Barry Jarrett has pointed out that there is an "internal reference junction" inside of PIDs, or at least inside of the Fuji PXR3, and if this is not ventilated to outside air the behavior of the PID controller can be erratic.

Maybe yes maybe no, depends on the machine and where mounted internally. PID can be safely mounted inside on some machines, and previously stated as such and why for the Bricoletta. Specifically yes PIDs have a max rated ambient operating temp of ~120f, so should be mounted in location that will always be safely below 120f. And already stated the lower back internal area of the Bric' is only 89f and is therefore perfectly safe area for mounting a PID. The Bricoletta has excellent convection air-flow with the internal top inside of case ~40f hotter than the bottom.

Here's another issue; do you want to control your shot temps?

Already stated "Direct shot temp control is not my reason for considering PID or tighter deadband PID. HX surf is and will be HX surf regardless."

I have no intentions to attempt to run HX & boiler tuned to function like a single boiler machine then requiring raising boiler temp/pressure for steaming. A single boiler non-HX E61 machine would likely be better suited for that purpose! (or other single boiler machine with relatively thermally stable group. Or dual boiler machine of course) Like others I have no problem hitting any desired target shot temp < + or - 1f HX surfing, I use flash flush-n-go.

Edit: Which isn't to say there's anything wrong with someone choosing to attempt to tune their HX machine for a specific shot temp via boiler temp with corresponding low boiler pressure. Since I choose not to mount a PID externally with the Bric' changing boiler temps all the time to change shot temp that scenario would be impractical. It may or may not work as effectively on a more common prosumer 1.5L boiler size range anyway, for someone else to experiment with!

[Ken Fox]

Maybe yes maybe no, depends on the machine and where mounted internally. PID can be safely mounted inside on some machines, and previously stated as such and why for the Bricoletta. Specifically yes PIDs have a max rated ambient operating temp of ~120f, so should be mounted in location that will always be safely below 120f. And already stated the lower back internal area of the Bric' is only 89f and is therefore perfectly safe area for mounting a PID. The Bricoletta has excellent convection air-flow with the internal top inside of case ~40f hotter than the bottom.

The issue I raised has nothing to do with the operating temperature, it has to do with ventilation. It is Barry's opinion based on his own experience that the controller must be ventilated to the outside and have through ventilation. The fact that a portion of the inside of your machine is below maximum stated operating temperature is not relevant to the issue of ventilation. I originally mounted my first PID in an external aluminum box which would have been at around ambient room temperature. After discussing this with Barry I was convinced that this was not satisfactory; there needed to be holes drilled in the project box itself even though there was no exposure to extreme temperatures.

Although my post attempted to address issues you raised, it was not a point on point response to your initial post, therefore the fact that failed to respond exactly to your questions and failed to take into consideration exactly what you were seeking is not because I had not read your original post. It was my opinion at the time that I posted that you had already made up your mind in favor of the pstat, therefore my response was not really intended for you, but rather for others who may read the thread either in realtime or in the future. Had I been responding specifically to your post, I would have quoted it in my post, and it is obvious that I did not do so.

ken