iSpindle - DIY Electronic Hydrometer

[pocketmon]

Not "all good". It only sends data to Ubidots when the reset button is pressed.

More investigation required.

You might have the same issue as Klooid.

Try connecting D1 to RST directly, but be careful not using USB. Sam said he fried his D1 that way.
The better solution is a diode.

 

[mattrox]

You might have the same issue as Klooid.

Try connecting D1 to RST directly, but be careful not using USB. Sam said he fried his D1 that way.
The better solution is a diode.

I think you are correct. I also think I may have a schottky diode on hand.

 

[Mikmonken]

V2 boards ordered fingers crossed these one will work, just another couple of weeks to wait ...

 

[Mikmonken]

Would the connection between the RST and D0 explain why when I to access the set up page via the reset button it takes a few goes to be able to access it?

 

[pocketmon]

Would the connection between the RST and D0 explain why when I to access the set up page via the reset button it takes a few goes to be able to access it?

I don't think so, but my knowledge about circuit voltage, current, and delay is very limited. Therefore, don't take my word for it.

Here is a piece of comment directly from iSpindel. Which might explain why you may need to reset it for multiple times to get the configuration page.

  // The ESP reset info is sill buggy. see http://www.esp8266.com/viewtopic.php?f=32&t=8411
  // The reset reason is "5" (woken from deep-sleep) in most cases (also after a power-cycle)
  // I added a single reset detection as workaround to enter the config-mode easier

 

[Klooid]

You might have the same issue as Klooid.

Try connecting D1 to RST directly, but be careful not using USB. Sam said he fried his D1 that way.
The better solution is a diode.

I soldered a 330ohm resistor to one of my boards and had no linkage with Ubidots. I have also a BAT43 Diode to try, and I'll keep you posted if that works.

 

[Klooid]

@Mikmonken: V2 boards ordered fingers crossed these one will work, just another couple of weeks to wait ..

Thanks Mikmonken, as I note that you have shared these on PBBs.io and they look good.

Once you give us the thumbs up, I'll follow up with my order.

Thanks again.

 

[Pronkster]

Hey guys, I've just upgraded my iSpindle based on the WEMOS D1 mini PRO with an external antenna. I had to re-solder the tiniest connector on the board which was very challenging to say the least but with pretty good results! The WIFI signal has improved about 50% to 60% which helps if your fermenting fridge is in the garage and 15 meters away from you WIFI router . Anyway, thought id share some pics

link to similar antenna as the one i am using: https://www.ebay.com.au/p/?iid=272609208646&chn=ps&var=571728694067

 

[Klooid]

I soldered a 330ohm resistor to one of my boards and had no linkage with Ubidots. I have also a BAT43 Diode to try, and I'll keep you posted if that works.

Just to let you all know that I built a new board and tried a BAT43 Schottky Diode, and my iSpindel is now working, which is brilliant!

Cheers

 

[gezzanet]

Cool

 

[kbaggen]

Hi there,
been folowing the forum a bit, but first post.

I went a bit other way for my "Alternative Ispindel". As I had a hard time to find a 3d-printer and aslo taking the time of printing into account I went another way.

I have used a liqued yeast tube and cut it Down a bit and placed it in the petling tube. The gyro is then placed at the yeast tube cap area!

The yeast tube also become a fine battery holder and by 2 springers from a battery cartridge I dont have to solder on the battery!

Some pics:

Only differncy to those pics, is I had to use 3 screws (se last Picture) insteed of only one to get the initial tilt of 23 in clean tap Water!

Hope someone else can use it!

Question, I decided to not remove light diode from gyro as I like to be able to see it is on! How does that interfluence the battery time? Iunderstood it would give around 3 months if removed, so what to expect on my 3200mha?

Comments welcome!

KBN
posebryg.dk

 

[crane]

Question, I decided to not remove light diode from gyro as I like to be able to see it is on! How does that interfluence the battery time? Iunderstood it would give around 3 months if removed, so what to expect on my 3200mha?

Comments welcome!

KBN
posebryg.dk

90 days is 2160 hours. This would mean that it is consuming an average current of 1.48mA without the LED for a 3200mAh battery(3200mAh / 2160 h =1.48mA). LEDs typically use 10mA, so your average current would be 11.48mA with the LED. This would drop your battery life to 11.6 days.

 

[rocketpants]

Question, I decided to not remove light diode from gyro as I like to be able to see it is on! How does that interfluence the battery time? Iunderstood it would give around 3 months if removed, so what to expect on my 3200mha?

Comments welcome!

I found with the gyro led still on that I could get ~14 days from a 2200 battery with 15 minute updates. I assume that if I change the 15 minutes to 30 minutes I would be looking at 20+ days lifetime from that battery

 

[kbaggen]

Btw.......the keen eye will notice I have the balacing screws on the same side as the Gyro on the above Pictures, and as such that is a wrong install as the gyro then will face the wort side. More corect according to univerSam is the other way. IT Works, but software 5.x.x "zero calibration" will not Work unless correctly install......so I am changing it by now wich give the issues the battery conenction have been solder on and of to many time, so have to just use "+" and "-" next by.......!

 

[ElChangoGuero]

Btw.......the keen eye will notice I have the balacing screws on the same side as the Gyro on the above Pictures, and as such that is a wrong install as the gyro then will face the wort side. More corect according to univerSam is the other way. IT Works, but software 5.x.x "zero calibration" will not Work unless correctly install......so I am changing it by now wich give the issues the battery conenction have been solder on and of to many time, so have to just use "+" and "-" next by.......!

interesting...where did you find that info? i cant seem to find it, but zero calibration does sounds pretty nice

 

[kbaggen]

I private mailed unveriSAM on hobbybauer.de, eg. he inventer of the Ispindel

Hi,

interesting version you made!

The orientation of the GY is quite critical since the internal calcs but even more the zero compensation is related to a well known dependency.
Therefore, it is not important where actually it is positioned but the orientation must not change.

See the pictures in the repo how it is set, basically upwards.

The DS you can position as you like, I set it facing downwards into the wort.

Hope that helps!
Sam

My question was:

Hallo Universam,
I am writting to you in English as I have one importan question in regards of the Ispindel!

As I had the idea of making it from old yeast tube placed inside of the petling, where PCB then is secured in the yeast tube also holding the battery, then my question goes around what is the most correct way of placing the gyro? Up or Down to the wort? Correctly I had it planed for being up and hence Away from the wort (hence the temp. probe get closer to wort!)?

 

[kbaggen]

Regarding the initial tilt.......as I had put the gyro Down towards the wort and hence a wrong install I re-did the setup, but apprently it change balance a good deal Now the initial tilt was only 12-13, and more weight drowned it!

But made a sugar-range-dilution measument!

And if you look at the figure you will se the curve is none-linear if we are below the tilt of 20, or said it other Words, you more or less can make one linear curve from tilt 0-20, and then another linaer from Tilt 20-90!

So to get a more linear approach the intial tilt should be over 20 the way I understand it!

 

[pocketmon]

Regarding the initial tilt.......as I had put the gyro Down towards the wort and hence a wrong install I re-did the setup, but apprently it change balance a good deal Now the initial tilt was only 12-13, and more weight drowned it!

But made a sugar-range-dilution measument!

And if you look at the figure you will se the curve is none-linear if we are below the tilt of 20, or said it other Words, you more or less can make one linear curve from tilt 0-20, and then another linaer from Tilt 20-90!

So to get a more linear approach the intial tilt should be over 20 the way I understand it!

my .002.

Sensitivity is the reason for initial tilt of around 20. The more difference of tilt degree, the better resolution we will get. However, we will have no control of which direction of tilt if the initial degree is 0.

I prefer cubic formula over quadratic formula. I got better approximation by having x^3 term.

You can use my calibration HTML file to check the 'accuracy'(or error) of cubic and quadratic formula to see which is better for you.
https://github.com/vitotai/BrewPiLess/blob/master/extra/calibrationSG.htm

The new version of iSpindel supports customized formula, so you can just use the one that is better.

BTW, it seems hat a few people have made their iSpindel. How do your iSpindels perform? Mine is very accurate, like most 1 point off.

 

[ElChangoGuero]

my .002.

Sensitivity is the reason for initial tilt of around 20. The more difference of tilt degree, the better resolution we will get. However, we will have no control of which direction of tilt if the initial degree is 0.

I prefer cubic formula over quadratic formula. I got better approximation by having x^3 term.

You can use my calibration HTML file to check the 'accuracy'(or error) of cubic and quadratic formula to see which is better for you.
https://github.com/vitotai/BrewPiLess/blob/master/extra/calibrationSG.htm

The new version of iSpindel supports customized formula, so you can just use the one that is better.

BTW, it seems hat a few people have made their iSpindel. How do your iSpindels perform? Mine is very accurate, like most 1 point off.

its seems to work just fine but havent actually done a fermentation with it, next batch for sure ill throw it in there

 

[Mikmonken]

my .002.

Sensitivity is the reason for initial tilt of around 20. The more difference of tilt degree, the better resolution we will get. However, we will have no control of which direction of tilt if the initial degree is 0.

I prefer cubic formula over quadratic formula. I got better approximation by having x^3 term.

You can use my calibration HTML file to check the 'accuracy'(or error) of cubic and quadratic formula to see which is better for you.
https://github.com/vitotai/BrewPiLess/blob/master/extra/calibrationSG.htm

The new version of iSpindel supports customized formula, so you can just use the one that is better.

BTW, it seems hat a few people have made their iSpindel. How do your iSpindels perform? Mine is very accurate, like most 1 point off.

Mine was a few points out from the hydromet of but I'm waiting till me boards come through before I use it proper as every time I take the cap off to charge the angle changes a few degrees as it's not fixed in yet.

 

[pocketmon]

Mine was a few points out from the hydromet of but I'm waiting till me boards come through before I use it proper as every time I take the cap off to charge the angle changes a few degrees as it's not fixed in yet.

That happened to me in my first usage, and I had to re-calibrate it.

I am thinking about gluing the sled to make it fixed, so that the sled won't be pulled out when pulling off the usb charger. A better solution might be wireless charge?

 

[kbaggen]

Added weight more at the top, and got initial tilt of 22-23! Seems to get better resolution at the low level then, but trade off is then low resolution at begging of fermentation at high OGs! But would anyway always take a OG measument!

Guess it is ready for a brew the next days........!

 

[JayEff]

A better solution might be wireless charge?

It's probably possible, but to make the energy transfer efficient, you'll need properties of your iSpindle to change, first and foremost, a cylindrical design will likely make the power transfer extremely inefficient. The coils need to be rather flat for example. Here's a great video on the subject, where someone builds the circuitry and coils by hand: [ame]https://www.youtube.com/watch?v=3E5PUnYlaTM[/ame] If you can adapt that, though, it would be a really cool modification!

 

[pocketmon]

It's probably possible, but to make the energy transfer efficient, you'll need properties of your iSpindle to change, first and foremost, a cylindrical design will likely make the power transfer extremely inefficient. The coils need to be rather flat for example. Here's a great video on the subject, where someone builds the circuitry and coils by hand: https://www.youtube.com/watch?v=3E5PUnYlaTM If you can adapt that, though, it would be a really cool modification!

How about this one? It's small enough to be placed at the cap where is the biggest flat area.

38mm OD is tight-fit.
https://www.aliexpress.com/item/Fre...r-Supply-Coil-For-Cell-Phone/32579790408.html
30mm should be fine.
http://www.ebay.com/itm/Wireless-Ch...923608?hash=item2a49de2c98:g:13QAAOSwimdXox0b
and even smaller 10mm
http://www.ebay.com/itm/1set-Wirele...821725?hash=item1e98b2025d:g:VwIAAOSwxYxUwjA5

 

[JayEff]

Interesting, I totally forgot about the cap but if that thing can transfer 500 mA I'd be surprised. It's possible, but I think that's the transmitter's current draw, not the transfer. But still, if even half of that goes through, that would be decent enough, charging the battery in theory in 5 hours, in practice probably upwards of ... 8? Charge current for LiPos should drop when they reach the max voltage and the charge protection circuitry will likely steal some current too.

For that price I'd say it's worth giving it a shot!

 

[Mikmonken]

Would you not still want to remove the lid to ensure the ispindel had been properly cleaned?

 

[pocketmon]

Would you not still want to remove the lid to ensure the ispindel had been properly cleaned?

Therefore, the coil should not be attached or glued to the cap. I don't have any plan for now, just thinking about it.

 

[Mikmonken]

Has anyone had any issues with the auto calibration? I get the light blue and then after a couple of minutes get and error and the light stays solid so assumingly not having calibrated successfully. View attachment ImageUploadedByHome Brew1495807679.398018.jpgView attachment ImageUploadedByHome Brew1495807697.853286.jpg

 

[kbaggen]

Hi agian,
made a US porter of the leftover yesterday and today! Got G=1054 by hydrometer, and Ipsindel says 1060!

Can be followed on:
https://app.ubidots.com/ubi/public/getdashboard/page/HipoQxyH7NhTHfugz6wIYfvMSDY

I assume from my curve that I get good precision at low gravity, but uncertain from high Gravity as the tilt was very low in sepration contra hydrometer readings at the high gravity readings over G=1050-1060! But as I would always take an initial gravity reading I guess that is the way to go?

 

[Klooid]

@Mikmonken, I found that using my iPhone or iPad was much better than the MacBook for connecting to and calibrate the iSpindel?

Also, I trust that your V2 Boards are not far away by now?

 

[Mikmonken]

Pcbs have been in fabrication for a little while now so hopefully will be sent out soon. I'll give the calibration a shot on my iPad tomorrow and see if I have a bit more luck. It goes into calibration but then seems to time out.

 

[Scarabrew]

Pcbs have been in fabrication for a little while now so hopefully will be sent out soon. I'll give the calibration a shot on my iPad tomorrow and see if I have a bit more luck. It goes into calibration but then seems to time out.

Are you planning to sell these PCBs? Because considering I don't have the expertise in soldering I would definitely buy one of these.

 

[pocketmon]

Has anyone had any issues with the auto calibration? I get the light blue and then after a couple of minutes get and error and the light stays solid so assumingly not having calibrated successfully.

It might already be done. Sometimes, ESP8266 can't restart successfully by software reset( watch dog reset, to be precise.).
I would suggest you to do it again. If it still happen, just move on and get the formula. It should be fine. I forget to calibrate the sensor, but I don't want to redo the whole sugar dilution process again. My iSpindel works very fine by reporting around 1 point errors.

 

[cerberos]

If anyone's looking for weights, try your local car tyre place, mine gave me these for free.

The v5 calibration function worked ok for me.

This link has a table detailing how much sugar/water/gravity for anyone doing the quick calibration method. https://mpesgens.home.xs4all.nl/thwp/sugar.html

 

[BrunDog]

Just make sure they are not lead!

 

[Mikmonken]

Are you planning to sell these PCBs? Because considering I don't have the expertise in soldering I would definitely buy one of these.

I wasn't planning on selling, the idea was that they would be very simple to hook everything up. Also based in the U.K. It probably wouldn't be very viable to send an assembled board in the post.

 

[Scarabrew]

I wasn't planning on selling, the idea was that they would be very simple to hook everything up. Also based in the U.K. It probably wouldn't be very viable to send an assembled board in the post.

Well, I understand. If for any reason you have any 100% working spare board I would be glad to buy it.

 

[cerberos]

I wasn't planning on selling, the idea was that they would be very simple to hook everything up. Also based in the U.K. It probably wouldn't be very viable to send an assembled board in the post.

I was thinking of trying to get some boards made up but first I'd have to learn out how to use a PCB design program (one that works on a mac). I'd love a copy of the Gerber file (or other format) if you're happy to share it.

Alternatively I'd like to buy a bare board from you (I'm UK based too).

 

[buckshot13]

Hi All,

I have just updated the iSpendel firmware to 5.1.2 and everything was ok. I logged on to Ubidots, and the divine came up straight away with all the variables on display. However, although the Wemos is flashing the intermittent blue light, the Ubidots is stationary with the first values and even after 30 minutes the values have not changes. I tried re-stating the device, and the Ubidots refreshes with new vales. Therefore, can anyone shed some light on what I done wrong please?

Thanks.

PS The setting at configuration is to send every 900 seconds (15 minutes).

Hello,

I also have exactly the same problem.
I've already set it to different update values (20sec. to 15min.)
Now I've also changed the 470ohm resistor with a schottky diode (ring to D0) and still the same problem.
Can anyone give some more help?

 

[dyqik]

my .002.

Sensitivity is the reason for initial tilt of around 20. The more difference of tilt degree, the better resolution we will get. However, we will have no control of which direction of tilt if the initial degree is 0.

I prefer cubic formula over quadratic formula. I got better approximation by having x^3 term.

You can use my calibration HTML file to check the 'accuracy'(or error) of cubic and quadratic formula to see which is better for you.
https://github.com/vitotai/BrewPiLess/blob/master/extra/calibrationSG.htm

The new version of iSpindel supports customized formula, so you can just use the one that is better.

Shouldn't the formula be a trigonometric one that you happen to be approximating with a polynomial? Since we're turning moment arms into angles. A cubic is probably close enough, but you might get a fit with fewer parameters in the model that way.

I should probably work out the physics, since I am a physicist...