Looking for a meter to go with E61 thermocouple adaptor

[dsc]

Hi everyone

I'm planning to buy Eric's TC adapter and wanted to couple it with a TC and a panel meter. The machine I want to install it in is an Andreja Premium and I plan to cut out a hole in the front case to put the meter inside. I wanted to keep the cost down, but unfortunately it's quite hard to do, well at least it's harder than I thought.

The first thing is the TC which costs around 30$. I need to plug it somewhere, so I thought of buying a panel meter, thinking that it would be cheaper than a regulator. Alas it's not, so I'm pretty much looking for a panel meter/regulator with a TC input. The main problem with cheaper PIDs/meters is their accuracy and resolution. I would love to have something with 0.1*C resolution and 0.2-0.3% error, but such equip. costs a fortune. The cheaper ones only offer 0.1*C resolution for RTD's like Pt100 or Cu50 which will be too slow to use on the group (although most offer 0.2-0.3% errors). On the other hand I'm not so sure if I need that 0.1*C resolution, as errors might be as high as 1*C.

So the main question is, what would you suggest for under 80-100$ (around 50GBP as I live in the UK) ?

Would love to here your ideas and thoughts:)

Cheers,
dsc.

[starry]

I believe that the inexpensive PID from Auberins will resolve to your requirement if you use a platinum thermocouple. They sell one for $10 and the PID is about $35.

[jesawdy]

I'd suggest sourcing the TC from Eric if he has what you need and the Auber Instruments unit is a good suggestion. The 1/32 DIN PID is $35 and can display 0.1C or 0.1F. You could also source a used PID from eBay and just use it as a temperature display unit.

[dsc]

Hi

well just to make things clear:

- Pt100 and Cu50 are RTD's (resistance temperature detectors) which are very accurate but slower than TCs

- TCs (thermocouples) are very fast but need additional compensation and cheaper PIDs don't offer 0.1*C resolution (I think because of the noise which would make the resolution useless anyway)

- thermistors, which are kind of similar to RTD's but are based on semi conductivity, quite fast, quite accurate and might be the thing to use

I'm still not sure which one I want. I was pretty confident to go with a TC, but now I'm not so happy about spending 150$ for a meter which will give me the option to have 0.1*C resolution.

Cheers,
dsc.

PS. that Auber PID offers 0.1*C only when using a Pt100. For TCs it's still 1*C only. Actually I've emailed them asking a question about resolution and they suggested using a thermistor. Anyone here who knows a lot about temperature sensors?

[BradS]

Here's another decent reference for comparing sensors:
http://www.controleng.com/article/CA6434738.html

[cannonfodder]

I use a Fluke 54 II. I know you are looking for a panel mount option, but after a few weeks of use you will not need the adapter any longer. Nice thing about the Fluke, I can use a Scace and the group adapter at the same time.

[erics]

The most expensive part of either the thermocouple adaptor kit OR the digital thermometer adaptor kit is the adaptor (and its associated parts). One aspect of the adaptor is to "get you inside your machine" in a fairly easy manner.

A solution for you would be to use the digital thermometer adaptor and purchase the thermocouple adaptor kit ($4) for possible use later on. Now true, you would be limited to a 1/8" probe (whether it be a thermocouple, an RTD, or a thermistor) but you can have a lot of worthwhile fun with the thermometer and take your time in researching the various other temperature measuring options available (should you determine that they would be better suited FOR YOU). Not only that but you can also ponder the surgery you describe on Andreja's casework which, IMO, you MIGHT regret later on.

More info available here: http://users.rcn.com/erics/

[dsc]

Hi all

It looks like I'm going to buy a new PID regulator which normally costs around 150$, but I might pay only 40$:) Will see how the auction goes.

Now the next problem is fitting the damn thermocouple in the group. Normally I would buy Eric's adapter, but because I'm on a tight budget I would love to cut costs a bit. I found similar TC adapters on Swagelok (Erics original idea) and Omega (OMEGAlok fittings) which both cost around 10GBP (but shipping for Swagelok costs a whooping 8 GBP, the same as the fitting alone:|). Now as Eric already noticed (and most of you on HB) there's a problem with screwing the fitting in the group as the hole is in a hollow. I came up with and idea to custom make a thread converter which would look like this:



The only problem is I can't find anyone who would make one for me. I was hoping to get it done in Poland (I'm originally from Poland) but most lathe operators won't do small quantities (I only need one or two).

I can always buy a second group screw, drill it and use some epoxy to stick the TC in place. I wouldn't want to do it as it means the screw is stuck to the TC for good and because the TC costs 15GBP it would be cool to use it in the future without something stuck to it.

I know that someone on HB did drill a M6 screw and it worked. Was it you cannonfodder?

Cheers,
dsc.

[HB]

I know that someone on HB did drill a M6 screw and it worked. Was it you cannonfodder?

See Naked Truth of E61 Temperature Revealed.

[cannonfodder]

I know that someone on HB did drill a M6 screw and it worked. Was it you cannonfodder?

Cheers,
dsc.

Yes, it was me. I would not do it again though, pain to do.

[mgwolf]

If you're on a budget, get the complete Swagelok adapter with fitted digital thermometer from Chris' Coffee for $85 or $90. It takes 10 min. to put it into your E61 and looks MUCH better than a TC with a wire hanging out from it. It will be as accurate as you need for the job. Michael

[dsc]

Hi

well I'm still looking for a lathe operator to make that thread converter for me, but I'm in contact with a friend who might be able to shape something up. Will see how it goes.

I was also thinking of using M6 threaded welding tips which are copper and already have a pretty nice shape and would probably fit pretty well. I will have to redrill them as they usually only have a 1.2mm dia hole max.

As for the thermometer Michael, why do you think it will be as accurate and fast as a TC coupled with a PID that has cold junction compensation and can display temperature with 0.1*C resolution? I seriously doubt it will be better and with used equipment I can get even below the 80-90$ budget, which is great news for me.

Cheers,
dsc.

[Richard]

It will be as accurate as you need for the job.

. . . why do you think it will be as accurate and fast as a TC coupled with a PID . . . ?

Michael did not say it will be as accurate; he said it will be as accurate as you need for the job. Those are different standards.

[erics]

The thermometers (mine) are based on either the Taylor 3516 digital (degrees F) or Taylor 9841 (degrees F or C - switchable). Both of these thermometers are currently "powered" by a 50K ohm thermistor. As received by me, the thermistors are inside a 0.151" OD tube, pretty close to the tip. When I am done with my modifications, the same thermistor is inside a 0.125" OD tube, very lightly forced into the tip, surrounded by Omegatherm 201 heat transfer grease.

The digital thermometers are "calibrated" in condensing steam and are tagged as such. They are surprisingly FAST. The time-constant for these little gizmos in condensing steam (going from room temperature to ~212F/100C is less than 1 second. The read rate of the thermometer is 1/s.

Now these numbers are not up to the standards of a similiarly sheathed type T thermocouple feeding a TYPICAL pid meter but you would be pleasantly surprised.

[dsc]

Hi

Richard you are correct, I shouldn't have stated that. But still I won't get 0.1*C and the response speed that I want with the thermistor thermometer.

Eric I believe you that the thermometer is fast, but not as fast as a TC and a good quality temp regulator/meter.

Just got a message from my friend that he will be able to produce that thread converter for me, so I should be able to complete the whole thing with just the basic elements, like the of-the-shelf TC adapter from Omega and some teflon tape.

Cheers,
dsc.

[cannonfodder]

Eric also makes a thermocouple version, which is what I use in combination with a ScaceII for testing.

[dsc]

Hi

what about 'naked' TC's? non sheathed type, that could be placed in a drilled screw and fitted in place with silicone?

I was also thinking of using this:

http://www.omega.co.uk/ppt/pptsc....ef=HSTC&nav=tema02

But I'm not sure how to fit it, besides that silicone + screw idea.

I think those partially sealed, or unsheathed TC's are going to be even faster, which for my application is just the thing I'm looking for.

So anyone tried using 'naked' thermocouples? I would probably go for a type T TC's, as type J has iron, which will corrode over time. That teflon TC's is also a good idea as it will never corrode.

Cheers,
dsc.

[dsc]

Hi again,

well here it is, my version of a TC fitting:



and here it is installed on the group:



I went with a used controller in a din 32 case:



The whole thing works very good, although when I flush the controller shows temperature of around 100*C. I also discovered that the temperature "on the screw" is around 4-5*C higher than that on the group.

I still have to solve a problem with flow speed and side channeling but that's a whole different story:)

Cheers,
dsc.

[CafSuperCharged]

dsc,

Your espresso machine looks like Andreja Premium.
Do you consider PID-ing it still (being HX and all) or will you just use the PID as a thermometer?

Peter
Netherlands
Europe

[dsc]

Hi Peter,

Yes it is an Andreja Premium. I will only use the controller as an indicator. In my opinion there's no sense in PIDing a HX machine. You could add an auto cooling flush mechanism, by sticking a TC in the HX and turning on the pump when it goes above a given level, but I don't think it's needed.

I might add a different solution to make the machine a bit more stable, but that's still only an idea.

Cheers,
dsc.