Barksdale 0.05bar pressurestat 1st look - Page 4

Need help with equipment usage or want to share your latest discovery?
Ken Fox

#31: Post by Ken Fox »

HB wrote:Given the inexpensive PIDs on the market, I agree that a PID / SSR combo is a viable alternative to a traditional pressurestat. And it may be that the results of your machine's unique group design and PID-controlled steam boiler will apply to machine's like Jon's, but I'm skeptical, especially after having installed Eric's TC adapter (for those who haven't heard about it, see Monitoring Brew Temperature - E61 & Silvia). Temperature surfing by altering the flush amount on the Junior simply doesn't work, but it works very well on the smaller HX machines. I bought an inexpensive PID from Eric. It includes a 1A SSR that could control the grouphead solenoid / pump on La Valentina to automate the flush with more precision than most have patience for by the "water dance" method or looking at a thermocouple readout. Now the minor problem of a free weekend to install it...
There are some issues here that your cryptic post smudges in a rhetorical way :mrgreen: , mon ami, Dan.

The fact that you find Cimbali Juniors to be not amenable to shot temperature surfing by flushing bears little relationship to whether some regimen of tight boiler temperature control and consistent flushing will produce consistent temperatures on OTHER machines. This would be, in my opinion, the sort of thing that would need to be tested and I'd enter the testing phase without any sort of preconceived notions as to whether or not it would "work." It could be, that for a given other machine, that a very much different regimen, as regards boiler temp and flush volume would produce consistent results. One would have to try a bunch of different scenarios in order to know if any of them "worked."

Just because variable flushes in a stock pstat-driven scenario in smaller HX machines seems to "work" doesn't mean that a boiler PID and consistent flushing wouldn't work also.

As to the cheap Chinese made PIDs one finds on ebay for a pittance; early reports indicate that they work ok, at least initially, but the comments I've made about PID controllers being nearly "bulletproof" apply only to those controllers that are commonly used in industry and hence have a track record. There are at least several brands of these, according to people who know much more about this stuff than I do. Fuji is only one in this group and not necessarily the "best."

In any event, I can't think of a good argument against being able to control boiler temperature easily, on the fly. The fact that machine purchasers have allowed mfrs. to get away with difficult to adjust machines (that have internal pstats) does not argue for functionality of this approach.

ken
What, me worry?

Alfred E. Neuman, 1955

User avatar
JonR10

#32: Post by JonR10 »

Ken Fox wrote:In any event, I can't think of a good argument against being able to control boiler temperature easily, on the fly. The fact that machine purchasers have allowed mfrs. to get away with difficult to adjust machines (that have internal pstats) does not argue for functionality of this approach.
On my machine in particular, the p-stat sets right under the top panel and is very easy-access but I do not make boiler pressure adjustments to vary brewing temps on the fly. I use a flush-and-go or flush-and-rebound routine to adjust brewing temp.

I agree that easily controlling boiler temperature is an advantage but there are reasons I'd prefer not to install a PID. I don't want to alter my machine in that way. Upgrading the p-stat still has the machine operate like a "stock" unit, but installing a PID changes the theory of operation (right?). Along the same line of thinking, I want to maintain a stock appearance, and having a digital electronic controller alters that. Lastly, I'd want to leave the PID inside the cabinet (so as not to show) and my machine is HOT inside so I would worry about ambient temps being harmful to the controller. As a note, I have not researched the allowable ambient temps for these devices so that may be a moot point.

User avatar
HB
Admin

#33: Post by HB »

Ken Fox wrote:Just because variable flushes in a stock pstat-driven scenario in smaller HX machines seems to "work" doesn't mean that a boiler PID and consistent flushing wouldn't work also.
To be clear, I agree that what you propose is worth trying and I would expect it to "work" since both approaches are forms of temperature surfing. My earlier point was simply that based on my experience with the Cimbali Junior, it comes as no surprise that variable flushes don't work, which is why we cannot assume your ample data will apply.
In any event, I can't think of a good argument against being able to control boiler temperature easily, on the fly. The fact that machine purchasers have allowed mfrs. to get away with difficult to adjust machines (that have internal pstats) does not argue for functionality of this approach.
Jon's point about aesthetics may play a part, and the general Italian manufacturer assumption that what's not broken doesn't need fixing. To be considered, any change would need to reduce their production costs while improving the machine's reliability. For those machines with a Sirai pressurestat (~$95 retail), an inexpensive PID / SSR combo might compete on price, but would probably lose on added assembly costs and potential warranty / serviceability issues. The "solution" is to demonstrate some added benefit (e.g., usability, flexibility, reliability) and then market it as a premium feature so they can pass the costs onto the (happily willing) buyer. Given the design of the Aurelia / Appia, it'd be a prime candidate for this strategy.
Dan Kehn

Java Man

#34: Post by Java Man »

Ken Fox wrote:Go for the long ball, Rick, PID the sucker :mrgreen:
I've been resisting the idea because I didn't want to clutter up my countertop with a project box. But reading what has been posted about various pressurestats -- including the commercial Sirai -- it seems that they're all problematic, one way or another. Diaphragms deteriorate even if the contacts don't. My only "over-pressure event" (OPE) occurred with my old M27 when the diaphragm on the Sirai crapped out. Paul Pratt posted on alt.coffee that he's had nothing but problems with Sirai, while another poster who put a Sirai in his Oscar reported that he was told by his retailer that the Sirais last ~ 1.5 years in commercial service. I only make ~ 3 shots per day, but I'd bet my small boiler element cycles on and off as much as a commercial machine used in an espresso bar. The risk of one OPE every 1.5 years doesn't thrill me. Surely a good PID must have a longer MTBF than that?

So . . . . about that PID . . .

I could put a project box on the top/back of my machine without cluttering up my coffee prep area, so why not? Is there any reason why a PID replacement for the pressurestat wouldn't control the boiler temperature better than a Sirai, and reduce the risk of OPEs??

I've also wondered about steaming performance, but I can't imagine why it would be any different from the way it is now. As soon as I crack open the steam valve on my Wega, the pressure drops ~ 0.3 bars, and it falls slowly throughout the steaming cycle. So even with a PID, shouldn't the element come on at full power and stay on as long as the steam valve is open?

I can't recall whether you've posted about how you did your PID replacement to your pressurestat. If so, can you post a link? Didn't you use a spare fitting on your boiler? (I don't have a spare -- I'd have to remove the pressurestat connection to the boiler.)

Cheers,

Rick
Java Man
A.K.A. Espressopithecus

Ken Fox

#35: Post by Ken Fox »

Java Man wrote:I've been resisting the idea because I didn't want to clutter up my countertop with a project box. But reading what has been posted about various pressurestats -- including the commercial Sirai -- it seems that they're all problematic, one way or another. Diaphrams deteriorate even if the contacts don't. My only "over-pressure event" (OPE) occurred with my old M27 when the diaphragm on the Sirai crapped out. Paul Pratt posted on alt.coffee that he's had nothing but problems with Sirai, while another poster who put a Sirai in his Oscar reported that he was told by his retailer that the Sirais last ~ 1.5 years in commercial service. I only make ~ 3 shots per day, but I'd bet my small boiler element cycles on and off as much as a commercial machine used in an espresso bar. The risk of one OPE every 1.5 years doesn't thrill me. Surely a good PID must have a longer MTBF than that?

So . . . . about that PID . . .

I could put a project box on the top/back of my machine without cluttering up my coffee prep area, so why not? Is there any reason why a PID replacement for the pressurestat wouldn't control the boiler temperature better than a Sirai, and reduce the risk of OPEs??

I've also wondered about steaming performance, but I can't imagine why it would be any different from the way it is now. As soon as I crack open the steam valve on my Wega, the pressure drops ~ 0.3 bars, and it falls slowly throughout the steaming cycle. So even with a PID, shouldn't the element come on at full power and stay on as long as the steam valve is open?

I can't recall whether you've posted about how you did your PID replacement to your pressurestat. If so, can you post a link? Didn't you use a spare fitting on your boiler? (I don't have a spare -- I'd have to remove the pressurestat connection to the boiler.)

Cheers,

Rick
Howdy Rickster,

Since I've done this TWICE, on two different machines, I've done it two different ways because the boilers, although the same size, are of different vintages and designs. The easiest one was the current vintage Cimbali Junior D, which has several unused ports easily used for TC probes. The 10 year old vibe Junior S, on the other hand, does not have a plethora of vacant ports and I, too, was reduced to using the port intended for the pressurestat (which is no longer in the machine).

I'm assuming that the Wega is like the Cimbalis in that there are a couple of other redundant methods of overpressure/overtemp safeties including a pop valve in the boiler and some sort of overtemperature element protection (a thermal fuse or thermal circuit breaker). In this case I think you will be just as "protected" with a PID failure as you would with a pressurestat failure, and safety is not being compromised. In the remotely possible situation where I am wrong, you would need to rethink the project.

First, you need a TC probe, and an ensheathed one would be the best. You could get someone to make one for you, you could make it yourself, or you could try to buy one from a source such as Omega Engineering. The fitting size you would need is obviously the same as the port where your pstat attaches to the boiler.

Once you have a probe, then you need to figure out how you are going to mount the thing. For my machine, an external mount with a small aluminum project box to house the controller was best. I used a project box slightly larger than the Fuji controller I used, and affixed the project box to the underside of the machine, being careful not to drill through any electronics or other critical wiring or plumbing or other components. This is the part of the project that takes the most time, to be sure that you won't damage the host machine and to position the controller so it won't get flooded should your drip tray overflow or other accident develop. Barry would suggest that in addition you should drill some holes through the sides of the project box so as to ventilate the controller, something that I have done myself.

Then you need to mount the SSR. If you do an underside of the case installation, then you can probably find another place on the underside that the SSR can be mounted on. The machine chassis can almost certainly dissipate the heat generated by the SSR without the need for a separate heat sink. This can be tested with a thermocouple and digital thermometer placed right up against the base plate. Mfrs. of SSRs publish charts of acceptable temps for their SSR baseplates and you can check to see that the heat is being adequately dissipated. This should be tested at the time when the SSR is going to be maximally stressed, i.e. when you heat the boiler up from room temperature and the SSR is in the "on" position for an extended period of time. If you do an installation like this it would be a good idea to invest $3 or $4 in a cheap plastic cover for the SSR; mouser.com and other electronics parts houses sell parts like this which vary by the brand and model of SSR you use. Again, you have to be careful that when you screw the SSR into the bottom chassis that you don't have screws going into anything important where they enter, and that the SSR is in a place where a water spill accident is unlikely to drain.

Once the controller is mounted and installed and the SSR is mounted and installed, and the TC is in place, you are left with wiring the whole thing up. You will need low amperage power to run the controller, which can be easily "stolen" from the electronics area or the incoming power line. You can use the wiring that powered the element from the old pstat to power the element from the hot side of the SSR. There will be 6 wires going into the controller (2 from the TC, 2 input wires for power from the electronics or input wiring, and 2 that go to control the low power side of the SSR), 2 from the controller to the SSR just previously mentioned, the pair of wires to complete the circuit to the element on the SSR's hot side, and that is basically it. A nice additional touch that may offer some protection would be to invest in a very low amperage (fraction of an amp) fuse on one leg of the power going into the controller, which will minimize the potential for "farking" (as Barry calls it) to other components in the event of a short. I bought a fuse carrier and fuse like this for about a dollar.

The "art" of the installation, that includes sourcing the parts and figuring out where to put them, will dwarf the time you will spend in actually wiring the thing up. Since I am not familiar with the Wega I can't suggest where to put the PID parts, but a little time spent on your part in examining the machine will be very well spent. There are a number of small aluminum boxes that will hold a 1/32 Din Controller such as the Fuji; I have bought three different ones that are each about the right size, with maybe an extra half to 3/4 of an inch in all dimensions.

If you want to get fancy you could also install a red LED light in parallel with the boiler element to show when the boiler element is being powered, and if you go overboard you could run a light parallel to the pump with your delay timer install (already done) to show when the pump kicks on. There is no end to the fun you can have :P

Best,

ken
Image

Image
What, me worry?

Alfred E. Neuman, 1955

Java Man

#36: Post by Java Man »

Thanks, Ken. This is a very helpful summary.

I also found a thread on AC in which you posted about the second PID installation and others added suggestions about wiring, fuses, etc. I'll be investigating this over the next few weeks to determine whether I can find all the required fittings and thread sizes.

Although my machine has a pop-valve on the boiler to prevent catastrophic boiler failure, I'd like to be able to find a PID controller with an "alarm" as suggested by Paul Monaghan so I could connect it to a relay that would kill the power to the boiler in the event of a system failure that leaves the heating element in the "on" setting. Maybe a little more complex and costly, but it would further reduce the risk of an OPE, and the pop-valve would become a semi-redundant last-ditch safety rather than the first backup.

I'll look for a project box that can house the PID controls as well as a temperature probe and readout for the E61 head as per Eric S's "kit". I may as well accommodate both projects in one box. There doesn't appear to be any rush to finish this since the spare MATER I've installed in the Wega is cycling reliably in a 0.95 to 1.05 bar band, with no apparent overshoots. I've had two MATERs fail, and in both cases, failure was preceded by a few months of gradually increasing deadband and occasional pressure excursions as high as 1.4 or 1.5 bars. I think I have time.

Cheers,

Rick
Java Man
A.K.A. Espressopithecus

Ken Fox

#37: Post by Ken Fox » replying to Java Man »

I don't think you can prevent each and every possible failure in a plumbed in espresso machine; you could get some sort of leak that would damage your wood floors when you went out for a few hours and many other things that are very rare . For that matter the icemaker in your refrigerator/freezer could have a leak and produce similar problems. But I digress.

I think that most PID controllers have an alarm circuit. What you would want to protect against, as far as I understand it, would be a failure of the SSR in the "on" position. If that happens, the alarm by itself is not going to shut off the power to the heating element; you would need additional wiring and hardware that I certainly do not understand at this point, and all of these things will take up what limited on machine space or inside the electronics box space if you have any room left there. Since a basic PID install is probably going to be at least as safe a method of controlling temperature as will a cheap pstat, I'd avoid desiging a system that is so complicated that in the end you will decide just not to bother with it. In my view the likelihood of this sort of (SSR/PID) "on" failure is sufficiently remote that I'd trust the secondary safety devices in the machine if that happened.

There is probably enough clearance under the machine (or could be if you adjusted the feet a little) to accommodate a small project box just slightly larger than a 1/32 DIN controller. If you want to add an additional display then you likely would need to put a larger box elsewhere, unless you could find a PID that would accept two inputs and display both of them. There are dual controllers but from what I have read they are very much more expensive and probably outside the range of what you would be willing to pay.

ken
What, me worry?

Alfred E. Neuman, 1955

Java Man

#38: Post by Java Man »

JonR10 wrote:Now, if anyone can find me a source for MALE BSPP adapters I would love to know about it.
Check out part # SS9035 at

http://mdmetric.com/prod/R631/StainlessSteel-pg1.htm

I don't know what sizes it comes in.

Cheers,

Rick
Java Man
A.K.A. Espressopithecus

User avatar
JonR10

#39: Post by JonR10 »

Java Man wrote:Check out part # SS9035 at

http://mdmetric.com/prod/R631/StainlessSteel-pg1.htm

I don't know what sizes it comes in.
Thanks again Rick,

Just curious, did you call them? It looks like those are made-to-order, and so the price for one piece would be astronomical and it would be a month before I could get one. I'm sure it would get much better if I ordered a few hundred (but it would still take a month).

In the meantime, I have already ordered a p-stat and the correct fitting from TerryZ at http://www.espressoparts.com

Java Man

#40: Post by Java Man »

JonR10 wrote:Just curious, did you call them? It looks like those are made-to-order, and so the price for one piece would be astronomical and it would be a month before I could get one. I'm sure it would get much better if I ordered a few hundred (but it would still take a month).
No, I didn't call because I'm now considering a PID replacement for the pressurestat. But I like your idea of ordering a few hundred. :wink:

I took the failed MATER apart, and it's obvious that in my machine, it's the electrical contacts that cause the short lifespan. I leave my machine on 7/24. I'm sure others may have scaling problems, too, and possibly diaphragm failures, but in ~ 18 months of use, the contacts in mine are burnt to a crisp.

This leaves another possible improvement open for those (like me) in a region with no water hardness problems -- installing a SSR to switch the heavy current, using the contacts in the MATER only for the control voltage. But I suspect I'll go all the way to a PID. I'll try to attach a photo of the MATER contacts.



Cheers,

Rick
Java Man
A.K.A. Espressopithecus