EScribe Software

[dwaindablane]

Hello Mr Evolv,

Any idea when the software will be released? itching to take a look at the goodies

[Phreys]

I'm also very interested! Can't wait to see if it's more "programming based" or "User friendly"

[John]

We are planning to put the download to it up some time tomorrow or early Monday.

[dwaindablane]

Cheers John, looking forward to taking a look at the software, looks amassing

[Phreys]

John said:

We are planning to put the download to it up some time tomorrow or early Monday.

Thanks John.  Was Escribe written from scratch by Evolv? Or is it based on other software I (or others) may be more fluent in?

[Brandon]

It was written in house.

[grizewald]

I hope so!

[macz]

Brandon said:

It was written in house.

Any plans for Mac version?

[John]

Not in the near term. We have a number of beta testers using parallels, which works reasonably well. All our new products from here on out will have this sort of functionality, so maybe a native Mac version in q4 or 2016, but it isn't on our firm development schedule yet.

[Rossum]

Virtual Box is free.    As is Bootcamp. 

[Alexander Mundy]

Will Escribe run on an NT platform? More specifically XP Pro or Server 2003? I am stuck in NT land due to $ of updating company software.

[Bapgood]

I believe the answer is yes to NT platform. However I personally haven't tested it. I just run it in compatibility mode for XP and it worked fine, but I don't know if that really tells us much. 

[SSV]

From a developers standpoint, what is the interfacing to this?  Is it JS in eclipse? or higher level?  I am interested in building discreet/custom wire profiles for our various wire products.  Is it going to be possible to build non-linear PTC scalars?  I.E. develop a TC scalar for each discreet product.  We have the necessary gear in our lab to test calibrations at various energy/heat levels.  I would like to give my customers a downloadable file for each of our wire compositions and sizes.  Thanks in advance.

[John]

You can build custom curves. No programming is necessary, they go in as csv files.

[dwaindablane]

This is looking like a more interesting board the with every post I'm seeing.

Kudos to John/Brandon and the team on a fine board, cannot wait to get my hands on the software.

[SSV]

John so if I export thermocouple data, plus resistance data tend curves straight to .CSV (my equipment can lock to sample clock so the trends are time correlated) I can just bring that trend straight into eScribe?  No hard scripting necessary?  That would be very nice.  Also can the software/firmware handle non linear points in the trend curve?  Our wire has some unique properties at saturation.....it transitions from a linear PTC to a sawtooth when it hits temp saturation.  So in essence I would like to have a calibrated scalar for each discreet wire product, that conforms to a reasonable degree of sDev accuracy across the coil surface.  I am assuming this can easily be done with a traditional PTC scalar measurement method?  Thanks, this is very good news indeed. 

[retird]

SSV, what wire are you talking about?

[SSV]

Our own wire products.....

[Bapgood]

SSV the CSV is just 2 columns, one for temperature and one for resistance. Or you can input the temperature coefficient of resistance.


 


 

[Alexander Mundy]

SSV, Maybe I am not understanding your post, but I don't see how a sawtooth could be implemented with simple resistance measurement TC. That would necessitate that the wire has a negative coefficient on the trailing edge of the sawtooth? How would the device know which side of the sawtooth it is on just by measuring resistance?

[SSV]

Alexander, this is why I asked.  You could track the tail (low side) of the sawtooth with a param for a "decrease" in resistance.....and implement a corrective lockout to NOT throttle power when that waveform is seen over X time.  We do this quite often in other thermistors, when they reach saturation maximums and become non linear....I have a lot of experience with dc/dc controllers and we typically do this through hysteresis

EDIT : here is a reasonably good article that might peak your interests https://sensorsandtransducers.wordpress.com/2012/02/07/thermistors/

because of the quantization of a 10 or 12 bit a/d the pure sine will be interpenetrated as a sawtooth.....in a 24 or 32 bit system we can get scalars that are more analogous to actual input waveforms, but I doubt it would be practical to implement a 24/32 bit a/d and associated vRef with any kind of stability in response....so we are stuck with interpolation errors, that look like a sawtooth

thankfully most modern McU have a hysteresis object library, to correct for such "problems".....simply patching a hysteresis control algorithm into the DO loop can help "smooth" the curve

[SSV]

BAP, that's perfect....the exact info I was looking for.....the typical curve for our wire is represented in this "basic graph.....please excuse the crudity, it's for reference purposes only

 

[Alexander Mundy]

I am fairly certain I read somewhere that it would only take increasing resistance values but I can't find the reference. I remember because I saw a curve one time that plateaued and went back down before increasing again and it made me think about what would happen in a closed loop system with simple resistance feedback.

[SSV]

Alexander, have a look at the last post I made to you....I amended it with some more information on how we can solve these potential "issues".....this is a fairly common subject we encounter, when dealing with thermistors, thermo couples, and RTD controllers


hysteresis is the name of the game =)

[SSV]

on a slightly whimsical, side note.....is there provision for bottom of the range scalars of sub zero temperatures.....you know just in case someone wants to vape accurately in the frozen tundra ?