PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-11-26, 03:42:17
News: Registration with the OUR forum is by admin approval.

Pages: 1 2 3 4 5 6 7 [8] 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ... 39
Author Topic: Itsu's workbench / placeholder.  (Read 137219 times)

Group: Professor
Hero Member
*****

Posts: 3498
Any problems to be foreseen with this 200V setup?
I don't foresee any problems with the circuit if the wire insulation withstands this voltage.

I used to use a flyswatter myself, but I have switched to an electric stun-gun (for self-defense) since then  >:-)

   

Group: Moderator
Hero Member
*****

Posts: 4159

Wow,  looking potent indeed, and was thinking of owning one myself, but they are illegal overhere now.
   

Group: Professor
Hero Member
*****

Posts: 3498
Wow,  looking potent indeed, and was thinking of owning one myself, but they are illegal over here now.
Any voltage is illegal in your country, or just above some limit?
   

Group: Moderator
Hero Member
*****

Posts: 4159

No,  any voltage is illegal, its not allowed to have one  and falls in the same category as firearms.
   

Group: Moderator
Hero Member
*****

Posts: 4159
While waiting for parts for the 45Mhz project (it takes weeks to get them nowadays), i dove into a home automation project i wanted to do.


Why?  Because here in The Netherlands we use mainly natural gas for heating and cooking because we have (had) a hugh natural gas pocket underneath the north-east part of our country.

As this gas is running out now and is causing earth quakes in that part, the governement decided to start pushing everybody off the gas and into alternative energy by slowly triple the tax on gas and halfing it on electricity.



So everbody is looking on his own way to heat his house and we decided to use the surplus of our 6kWh solar system (24 x 250W panels) for (extra) heating.

That solar system was 8 years ago calculated for the 6kWh because then we on average use about 5700kWh/year.
Since then, because of led lamps and other energy efficient matters, we now dropped to about 3500kWh/year.

This surplus of 2500kWh/year is now retuned to the provider at a third of the costs (7ct/kWh) we have to pay for it when buying from them (22ct/kWh).

So we have 2500kWh/y to spend on heating next to the gas system so we bought 2 infrared heating panels, one 1000W, and one 700W to aid in heating the livingroom first.


To get a good insight in the kWh uses and the hoped for drop in M³ Gas, the need for a good home automation system arose.


I decided for DOMOTICZ.

They have a starter kit which includes a Raspberry PI with their Domoticz program which has a mass of sensors already implemented.

It runs as a headless system, so without screen, keyboard and mouse and is reachable by its webserver on every PC, laptop, tabled or smartphone and is placed into the metercloset where all the energy meters are.

This starter kit comes with a P1 USB cable (interface) to interface with the smart meters (via their P1 port) to monitor Electricity and Gas.

Its kind of plug and play and within a few hours i had the real time (10sec intervall) data for Electricity uses and Gas uses with nice graphs etc.


The Electricity shows the 4 counters we have (2 for incoming energy (high and low tariff) and 2 for outgoing energy (high and low tariff).

But i still missed the yield of the solar panels to see what is effectively used directly (most cost efficient).
We do have a solar data monitor system, but its in the metercloset and needs to be scrolled through by pushing a button there.
It does also have a webpage where you can login to show your average yield/day, month, etc, but is kind of slow (hours).

This solar data monitor system box has a led that pulses 1000 times per kWh, so i decided to use that pulse to interface with the Domoticz Raspberry PI as it has a pulse counter interface.

I bought a 5-port pulse sensor USB cable which is also supported by Domoticz, build a pulse detector using a photo transistor followed by a 555 timer to shape up the pulse signal (30ms pulses needed) and an opto coupler to isolate it.
This pulse was fed into the 5-port interface and connected to the Raspberry.
It now shows very accuratly the real time (10s) yield in watts and overall yield/day, month, etc. of our solar system.

Comparing the Gas uses, Electricity uses and the solar yield is now very easy as it is displayed on 1 web page.

Finally not because it was needed, but because i can (still have 4 counter ports left), i decided to try to monitor the water uses.

It seems that my watermeter has a hidden hole (covered) in which one can put a hall sensor which senses a magnetic field from a magnet placed on the half of the last digit (0.1L) disk in this meter.
So every 0.5 liter the Hall switches on and the next 0.5L off.

Again i used a similar circuit to shape up the Hall sensor signal with a 555 timer, but followed it by a 74HC221 to only trigger on a UP going pulse from the Hall/555 (verpies design, slighly modified).

The result is a pulse (30ms) every liter of water and again is passed on to an opto coupler into the 2th port of the 5-port pulse counter.

After some adjustments, the Domoticz now shows the water uses/day, month, etc. in liters and M³ water.

All in all i am pleased with this result and it shows during the last month that we are on the right track by using the surplus solar yield and dropping the gas uses by about 50%.

Video here:  https://www.youtube.com/watch?v=mYESkq3ahlc

Dashboard picture here: 

Regards Itsu
   

Group: Professor
Hero Member
*****

Posts: 3498
This surplus of 2500kWh/year is now retuned to the provider at a third of the costs (7ct/kWh) we have to pay for it when buying from them (22ct/kWh).
Aaassssholeeess!
Anyway, do you know the breakdown between the cost of electric energy production and the cost of electric energy transmission, for your electric utility company ?

All in all i am pleased with this result and it shows during the last month that we are on the right track by using the surplus solar yield and droping the gas uses by about 50%.
I love it!  Did you have to do a lot of programming ?
It still has a room for improvement through selling the excess kWh to your neighbor.
Also, increasing the capacity of your hot water tank A LOT, insulating it with shiny aluminum foil + polyurethane foam and using the heat energy accumulated in it during the day for house-heating during the night, seems like a good idea, too.

   

Group: Moderator
Hero Member
*****

Posts: 4159

Quote
Anyway, do you know the breakdown between the cost of electric energy production and the cost of electric energy transmission, for your electric utility company ?

Well, no, not really, they keep it as vague as possible.


Quote
It still has a room for improvement through selling the excess kWh to your neighbor.

8 years ago we decided as neighbors to setup this solar system on our houses, so basically all neighbors have a similar system.
Perhaps i need to buy an electric car.

My gas unit has a small hot water tank for direct uses, but its in the gas unit, so no additional water tank yet, but of course its a good idea indeed.


Itsu   
   

Group: Professor
Hero Member
*****

Posts: 3498
Well, no, not really, they keep it as vague as possible.
In Czech Republic (also an EU country) the transmission of 1kWh costs 0.036EUR ( ...that does not include the energy generation cost, which is: 0.065EUR / kWh )
It might be useful to remember, that the energy transmission cost is approximately HALF of the energy generation cost.

So the electric utility company IS ripping you off (especially that day tariffs are more expensive than night tariffs), and that might create a business opportunity for a lawyer, in which he knocks on the door of every house with solar panels and collects signatures for a class-action suit.


P.S.
Did you have to do a lot of programming to integrate it all together?
   

Group: Administrator
Hero Member
*****

Posts: 3960


Buy me some coffee
Well done Itsu

Sounds amazing, i capture rain water and process it through a RO unit to produce pure water for my marine fish tank, the waste water goes onto another water butt to feed the greenhouse watering system and does have the possibility to feed my toilet flush (not yet plumbed it in though)

In the uk for every litre of water used from the mains they assume a percentage goes down the toilet and sink and you get another bill for waste water treatment based on your water usage.

I do have an electric car and could really do with installing solar PV, i used 7552kWh of electricity in the last 12 months, this is including my commute of 46miles to work and back each working day (about 14kwh a day), for that amount of electricity i paid £1467
   

Group: Moderator
Hero Member
*****

Posts: 4159
In Czech Republic (also an EU country) the transmission of 1kWh costs 0.036EUR ( ...that does not include the energy generation cost, which is: 0.065EUR / kWh )
It might be useful to remember, that the energy transmission cost is approximately HALF of the energy generation cost.

So the electric utility company IS ripping you off (especially that day tariffs are more expensive than night tariffs), and that might create a business opportunity for a lawyer, in which he knocks on the door of every house with solar panels and collects signatures for a class-action suit.


P.S.
Did you have to do a lot of programming to integrate it all together?


In my yearly specification, i read nothing on energy generation costs.
There is something they call delivery costs (probably your transmission costs) which went up from 11.4 ct/day in 2018 to 12 ct/day this year. (could be it has the generation costs included here).

Further they juggle with tariffs per day or per kWh, before tax, with tax etc. so almost impossible to understand.

Anyway, at the bottom of the specification is the amount to beat which was successfull the last 8 years (no electricity costs and some €250 / year returned).
 
 
Quote
Did you have to do a lot of programming to integrate it all together?

I tried to avoid doing any additional programming like using LUA scripts or Python scripts which seems to be common to use in adding sensors, so therefor i went for Domoticz which has a vast array of sensors already in it.

So no, no additional programming needed, i only needed to change the Raspberry PI DHCP assigned IP-address into a static IP-address in my network so i can easily browse to it (done via putty on Windows).

Itsu

« Last Edit: 2019-03-06, 10:28:44 by Itsu »
   

Group: Moderator
Hero Member
*****

Posts: 4159
Well done Itsu

Sounds amazing, i capture rain water and process it through a RO unit to produce pure water for my marine fish tank, the waste water goes onto another water butt to feed the greenhouse watering system and does have the possibility to feed my toilet flush (not yet plumbed it in though)

In the uk for every litre of water used from the mains they assume a percentage goes down the toilet and sink and you get another bill for waste water treatment based on your water usage.

I do have an electric car and could really do with installing solar PV, i used 7552kWh of electricity in the last 12 months, this is including my commute of 46miles to work and back each working day (about 14kwh a day), for that amount of electricity i paid £1467


Thanks Peter,

Allthough not intended to monitor, the water sensor shows some amazing things like every toilet flush is about 8 liters of water, the amount of water used by the dish washer, the washing machine or for a shower etc.


My milage is very low (46000km in the 12 years on my present car), so allthough an electric car was planned in combination with the solar system, it never came of it, not economical enough, perhaps in the next years.

Your costs are about 19 ct (penny?) / kWh then, which seems similar as we have.


Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4159

I bought myself a external reference 10Mhz oscillator which is linked to the GPS satelite system for optimum accuratie.

This GPSDO (GPS Disciplined Oscillator) has an Oven Controlled Crystal Oscillator (OCXO) which is synced to this GPS system via an antenna.

The meaning is to use this highly accurate reference 10Mhz signal to sync my Spectrum Analyzer (SA) and Function Generator (FG) with (and lateron perhaps my Ham radio stuff).

It seems that by using this setup, my tracking generator in my SA now has the same stability as this external reference signal , same deal with the FG.

The FG sometimes reject the presented reference signal and switches back to its internal oscillator, so there something needs to be checked.

   
Picture of the box see below, a video of some testing here:

https://www.youtube.com/watch?v=s1_al6vIvF4


Regards Itsu
   

Group: Professor
Hero Member
*****

Posts: 3498
In the "Specifications" chapter and the "Internal/ External Reference" section of the DSA815 Manual, it is written, that the maximum amplitude at the ExtRef input should be +10dBm, which at the 50Ω input impedance translates to 2VP-P

So, if the GPSDO is maintaining the 5VP-P at 50Ω load then you are overdriving the DSA815 quite a lot.  Measuring the amplitude of the GPSDO at the end of a BNC cable terminated with a Tee and a 50Ω terminating resistor*, would give you more accurate amplitude reading.

Also, the DSA815's Clock Reference output has 50Ω impedance while the DG4102's ExtRef input has 1kΩ impedance** , so without a matching termination*** at the end of the cable you are going to have terrible reflections!
After you eliminate these reflections and verify, that the external clocks' amplitudes are within acceptable limits (in-circuit with the Tee at inputs), you should run all of your tests again.


*   ...or scoped in-circuit with a Tee at the real 50Ω ExtRef input of the SA.
** see the DG4102's Manual, chapter "Specifications", section "Clock Reference", subsection "External Reference Input", point "Impedance (typical)"
*** 53Ω in parallel with 1kΩ  yields 50Ω. So a 53Ω BNC "feed-thru" termination resistor would be ideal to match the SA's clock output to the FG's ExtRef input.
   

Group: Moderator
Hero Member
*****

Posts: 4159

hmmm,  i would have expected that equipment of the same manufacturer would have the same specifications
regarding common things like these external reference ports.

Thanks for pointing out the opposite.

Probably 2 (or more) teams working only on either FG's, Scopes, SA's etc. like at my former employer.

Diving into it....Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4159

A quick check shows that the output of the GPSDO at the end of a short coax cable which has a tee with a 50 Ohm
terminator on 1 side is 3vpp just like the specs of the GPSDO mentions.

So far so good, no overloading of the SA.

Will deal with the FG mismatch later today.

Itsu
   

Group: Professor
Hero Member
*****

Posts: 3498
A quick check shows that the output of the GPSDO at the end of a short coax cable which has a tee with a 50 Ohm terminator on 1 side is 3vpp just like the specs of the GPSDO mentions.
So far so good, no overloading of the SA.
Are you sure?
3VP-P translates to +14.5dBm at 50Ω...and the DSA815 Manual states, that the maximum level at the External Reference input is +10dBm

P.S.
I am curious how accurate your VC3165 frequency counter is, too.
   

Group: Moderator
Hero Member
*****

Posts: 4159

I meant that the GPSDO is working according to its specs (3vpp output),  but yes its still to much for the SA.
Will insert a 3 or 6 dB attenuator inbetween.

The FG still (only at first poweron) rejects the external ref signal, even now with a 50 Ohm resistor parallel.
After some time (hour) it accept it, not sure why that is.

The vc3165 was at 10Mhz about 20Hz off, so now i have it calibrated within 10Hz (last digit not visisble).

Itsu 
   

Group: Moderator
Hero Member
*****

Posts: 4159

3dB attenuation did the trick,  2Vpp in 50 Ohm into the SA.
The FG does not accept the signal with the 50 Ohm parallel to it, without it it does.
   

Group: Professor
Hero Member
*****

Posts: 3498
The FG does not accept the signal with the 50 Ohm parallel to it, without it it does.
What is the signal level at the FG's ExtRef input ?
The DG4102 Manual states, that is should be between 250mVP-P and 5VP-P... square.
   

Group: Moderator
Hero Member
*****

Posts: 4159

2.9Vpp when open, 2.8Vpp with 1K parallel and 1.5Vpp with 50 Ohm parallel

Just noticed that my coax jumper cables are 75 Ohm.
   

Group: Professor
Hero Member
*****

Posts: 3498
Just noticed that my coax jumper cables are 75 Ohm.
Ooops!

P.S.
Did you notice that the FG outputs a square wave and expects a square wave at its 10MHz ExtRef input ? 
BTW: that input is AC Coupled, so a single-supply sine-to-square wave converter will work (e.g. 74VHC14)....but where to get the +2V to +5V supply voltage for it ?
   

Group: Moderator
Hero Member
*****

Posts: 4159


yes,  i will make some 50 Ohm jumpers first.

Indeed the FG external ref output is a 10 Mhz square wave DC signal 5Vpp.

When attaching the SA feedthrough 10Mhz signal and switching to external reference, the signal changes to a 2.9Vpp sine wave mostly above the zero line, so no pure AC sine wave.

So there is some battle at that single in/out ref port of the FG.

First i will switch to 50 Ohm jumpers, then see if i can transform the 10Mhz sine wave signal from the SA to a square wave DC signal to feed to the FG.


By the way, i have the RS232 port of the GPSDO connected to my PC, so i have accurate time and all kind of info from the GPS satellites circling above.


Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4159

I tried to make myself a sine to square wave converter using a 74AC14.

The circuit was simulated in LTspice (using a 74HC14) and shows what i want (picture 1).

After building this circuit using a 74AC14, i see some problems, like the distorted sine wave,
the ringing on the square wave, but the basis seems OK, see screenshot.

I use a 10Mhz signal (2Vpp @ 50 Ohm) from the FG for now.

Guess i need some better matching, see the 4.599v input sine wave while the FG was set to 2Vpp @ 50 Ohm.
The 5v supply is bypassed with 100nF ceramic and a 47uF cap.
All unused input points are grounded.

Itsu
   

Group: Professor
Hero Member
*****

Posts: 3498
I use a 10Mhz signal (2Vpp @ 50 Ohm) from the FG for now.
Good for testing.

Guess i need some better matching,
Yes, the input of the sine2sq converter should have an impedance of 50Ω. Sth like the circuit below might be a way to realize it.
Experiment with R1 and R2 down to 50Ω (but always R1=R2) in order to achieve minimum sinewave distortion at the input of the converter. Finally, adjust R3 for minimum squarewave distortion at the FG's ExtRef input, when the converter's output is driving it.

The 5v supply is bypassed with 100nF ceramic and a 47uF cap.
The 47μF might be an overkill at 10MHz. I would trade it for 1nF.
Bypass the VCC and Gnd in a similar manner, too, at the R1 & R2 connections.

P.S.
Avoid inverters (even Schmitt inverters) with TTL level inputs such as 74HCT14 and 74ACT14 and 74VHCT14 because the ½VCC voltage level is considered a HIGH level by their inputs.
« Last Edit: 2019-03-15, 22:09:01 by verpies »
   

Group: Moderator
Hero Member
*****

Posts: 4159

Thanks,

tried with 2x 100 Ohms and a 1nF ceramic instead of the 47uF, but no real change seen.
Will try tomorrow with lower values resistors.

Lowering the input signal to 1Mhz shows very nice signals, so this 10Mhz might be to high for this 74AC14 .


Itsu
   
Pages: 1 2 3 4 5 6 7 [8] 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ... 39
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-11-26, 03:42:17