My Motors got me to Tap into Radiant Energy

[Hitby13kw]

Hi,
thanks for hints
Those BT modules do not need any Arduino driver. They perform as fully transparent com port from Arduino serial port and perform far above usual PC COM 115000bit per second..

Thank You John


That explains why we couldn't find any driver ~ Don't need one . . . if that is true, it’s time to ‘put a fork in it’ we are done looking at individual modules ~ time to cook a meal.

John this is just a guess on my part (Please correct me if I’m wrong) but the job of the Arduino is just a traffic cop. Something on the line of:
read Temperature sensor 1 > send to BT- read ADC port A0 > send to BT - read Temperature sensor 2 > send to BT - read ADC port A1 > send to BT - read RPM sensor > send to BT – repeat

Hitby13kw

[Hitby13kw]

John, almost forgot to ask. Does your app want any particular string of ASCII characters between each sensor reading and does it want anything at the end and / or beginning to tell it:

Start of a new string setx - x - x - x - x end of set.

You know what I’m asking, does it need to conform to any given data protocol or are we setting up our own ?

Hitby13kw

[JohnStone]

Thank You John


That explains why we couldn't find any driver ~ Don't need one . . . if that is true, it’s time to ‘put a fork in it’ we are done looking at individual modules ~ time to cook a meal.

John this is just a guess on my part (Please correct me if I’m wrong) but the job of the Arduino is just a traffic cop. Something on the line of:
read Temperature sensor 1 > send to BT- read ADC port A0 > send to BT - read Temperature sensor 2 > send to BT - read ADC port A1 > send to BT - read RPM sensor > send to BT – repeat

Hitby13kw

Right!
Arduino will do perhaps some additional value conversions in order to get a readable output at PC as ASCII terminal (first step -> second se #4 below)
Exception: For rpm! Arduino will use his built in routine for frequency measurmeent or pulse cycle.
It is essential to get as many data we can get because then we are able to calculate true RMS more precisely.

1.
I imagine to perform procedure you described non stop but toggle an output bit at every start of data read. Then we can exactly measure how long one single procedure takes.
BTW: ADC can be red 860 times per second

2.
As we need an exact time sequence we start procedure by internal timer leaving some idle time inbetween procedure for some additional tasks. We need to have some time left after each sequence for some additional tasks. But for 1st step a time beat of 1second wil be reasonable.

3.
If we detect that there is additional performance we can load Arduino with RMS calculation. That is a bit tricky because we need to calculate RMS out of an array being operated like a shift register. Simple average calculation will not perform well. Let's look at this issue in 2nd step. I need to ponder on it.

4.
As soon we have agreed for a data format I will start writing a nice dashboard program along scope display and gauges for all we need to see. At PC we can do extensive calculations along data logging to EXCEL files.
Cheers
JS

[JohnStone]

John, almost forgot to ask. Does your app want any particular string of ASCII characters between each sensor reading and does it want anything at the end and / or beginning to tell it:

Start of a new string setx - x - x - x - x end of set.

You know what I’m asking, does it need to conform to any given data protocol or are we setting up our own ?

Hitby13kw

NOP!
We can transmit a well readable sequence,comma separated and CR at end.
It could be helpful if we start the string with # character and a string type ID 1...9.
We possible decide later on to transmit amps / volts much more frequently than rpm and temperature. Those deviations in format we will mark with #2 ...#3 .....
Then I can parse the string correctly.
JS

[JohnStone]

Instructable Section C: Opto input

OK Cornboy, you got your PSU running. Sorry for that forgotten wire jumper . But you took the chance to learn how a voltage regulator behaves if its GND pin is floating. You never will forget it !

Today you will assemble section C with opto.

First of all I apologize for some errors.

• Resistor R1 has proved to be 330Ohm
• At opto pin naming C and E are reversed in circuit diagram but pinning is connected OK and PCB is OK - no modification required.
• Those two wire jumpers are a mystery to me. It is copper below -> ignore them.

BTW: After you have your monsters running I will add this tutorials and corrections to my 5.1 document and will issue a final version V6.0.



OK hands on now:
1.
Before you assemble the section C we need to discuss what gen output you intend to use. For 5V R2 shall be 330Ohm (Arduino). The value of 1K was dedicated for 15 volt outputs like from NE555 timer driver.
Arduino outputs source current in high state and sink current in low state. Similar with lab generators at TTL output. But lab generators might not be able to source 10mA for opto.
If you use a normal generator output please make shure to be 50 Ohm output (and not 600Ohm) and adjust output voltage to oscillate between 0V and 5V (check with scope) Adjust frequency for about 1 Hz.

2.
Assemble section C
You might want to use an IC socket for opto - just at this first board (easy to preplace in case of damage). You need to cut those few pins out of a normal IC socket you have available.
You shall assemble IC socket for IC 4 but you shall not populate it just now. It will be more easy to measure at those unpopulated socket pins.

3.
Please regard that your generator owns a private GND. You need to change ground lead of your meter if you measure at generator side. Do not forget to restore GND if you contunue checkign your circuit.

In order to test your generator signal signal connect:

• Gen output (or Arduino) to K1 pin 5 (signal Rin)
• Gen GND to K1 pin 2 (YES pin two!!!!)
  ATTENTION: I insist in pin 2 and NOT pin 1. We are not allowed to connect Arduino to GND of monster driver.

Get an LED and connect its pins to K1 pin 4 and K1 pin 2. The LED shall be pulsating. If not reverse pins of LED. If still not get another being OK.
Disconnect test LED

3.
Connect K1 pin pair 3,4. Usually you solder a bubble between pins at bottom side for stable connection.

4.
Connect power to your monster board. Check 5V to be present at regulator output and opto IC3 pin4.

5.
I assume your generator (Arduino) is still connected with pulsing signal.
Check K1 pin 5 for pulsing signal (5V, 0V ......)
Check K1 pin 3 and 4 for same pulsing signal but now 1.6V ...0V .....

Explanation of opto function:
The opto input is a simple LED (IC3 pin 1,2). If you operate it without resistor in series it will die like a normal LED.
This LED emits light inside the sealed housing of IC3. At pins 3,4 there is connected a simple transistor. And now I explain the opto trick! This transitor has no base leg and under normal condition it would be not possible to control it. But the base of this transistor is uncovered and receives photons (light particles) from built in LED. The energy of these photons performs the miracle that our transitor decides to perform its switching function.

For more details see.

Imagine this transitor to be a simple switch. This switch is being toggled by "magic" synchronously with generator signal. What do you expect at IC3 pin 3 to happen? You switch 5V to a resistor ON / OFF / ON ....

• Check pin 4 to be stable 5V
• Check it at pin 3 to have pulses.
• Check IC4 pin 9 to have same signal
• Get your test LED and connect it to IC3 pin3,4. If it does not pulse reverse pins. LED will go OFF if opto being active (it short circuits test LED)

OK: That's it!
Next time we will discover the mystery of IC 4
JS

[JohnStone]

Instructable section D: Power on delay / signal conditioning

Today we will discover the magic of IC4. In fact the opto could drive the FET driver IC7 directly. Many circuits do it this way and it is OK - for low power circuits only.
In our case we intend to have a rugged high power driver and in this range any misalignment or uncertainity can destroy the circuit - in best case.

- Imagine you switch power of your driver on and without any control the motor gets a short kick and jumps from your desk - in best case. That is a very usual behaviour of electronic circuits. Thy operate OK AFTER they got settled. You might know this effect from switching on audio amplifiers. Some of them emit a loud BLOPP. We want to prevent those "BLOPP" effects in our setup.
The responsible circuit is D3, R4 along gate IC4 gate c

- Another effect needs to be prevented. Our FET driver is extremely fast in switching matters. In order to perform well it needs a very clear and steep input singal. Else it could perform some intermediate ON/OFF/ON..... actions called: bouncing. FETs might die because of different additional effects. Compared to FET the driver our opto is extremely lazy and can not perform the signal qualtiy required. Apart that I do not know what poor signal you feed in from your generator. But do not care! Any lazy and noisy singal slope will perform well because auf IC4. It is a so called Schmidt-Trigger model.

- You might asked yourself: Why do we use an extrmely fast FET driver and lazy opto as signal source. We need to look at two properties:
- Signal frequency is quite low in lower KHz range - no effort for whole circuit
- Switching speed: If a switch action is required - eventually - it needs to be extremely fast in order to keep FETs cool and foster radiant effects.

Of course if we would like to switch our monster driver with 100KHz - it will perform this frequency as well. But then we need an very fast and more expensive opto. But as this is not our focus we drive our 300HP monster with 30mph because we like its accelleration performance and not high speed.

......
OK: hands on now:

1.
Assemble Section D - except IC4
Rememeber to add C1 to section D

2.
Let's look at power on delay: D3, R4, C18 - Power ON delay
Imagine C18 to be a bucket. Your circuit will be enabled if this bucket is full of water (electrons = voltage).
Initially the bucket is empty. At power on R4 performs as water tap being somewhat opened filling the bucket with electrons. D3 is inactive because polarized in reverse direction.
After a predefined time the bucket c18 is full and asks gate IC4c to enable monster driver (high signal at input pin 10)
This gate is a so called AND gate (sorry for European representstion of symbol). The ball at output (pin 8) represents an additional signal inversion (negation). This combination is called NAND gate. Now let's look at its function.
Recall pin 9 is our pulsing signal.
- If both inputs are low (0V) the AND function gives low as well but because of negation it is high (5V)
- If one signal is high (5V) (recall pulsing signal at pin 9) the output will not move at all. (recall our bucket is low - initially)
- But if our bucket C18 is full (5V) it alows the NAND gate to move and the output will now emit our input signal - but reversed in polarity (0V at input -> 5V at output and vice versa)

Now: what about D3? It stays there upt to the moment we switch power off. Then 5V decrease rapidly and bucket C18 can be drained as fast in order to be prepared in empty state for nex power on sequence.

3.
- Switch on power at your PCB

Check IC4 pin 14 to be 5V
Check IC4 pin 7 to be 0V

-Switch power off

4.
- Assemble IC4
- Connect your generator with 1Hz signal
- Switch on power at your PCB

Measure IC4 pin 8, 12, 13 to be pulsing (0V, 5V)

5.
While your PCB running - short circuit C18

Measure IC4 pin 8, 12, 13 to be stable high (5V)
This would be the disable condition at very first milliseconds after switching power on.

6.
Connect your probe to C18 and switch power off. The voltage ar C18 shall go down to about 0V rapidly.

7.
Unfortunately output IC4 pin 8 is not what we require in order to get our driver quiet at power on. If we would connect IC4 pin 8 to our driver it would be constantly ON at power on. Therefore we used a second gate of same type. Both input pins are connected to output IC4 pin 8. This circuit performs as simple inverter. Remember we had an inversion before converting a logic AND function to logic NAND function. Now we add another negation. In sum we have a simple AND function as we require for our FET driver.

Measure IC4 pin 11 to toggle in same polarity like your generator signal.
(Please regard that your generator owns a private GND. You need to change ground lead of your meter if you measure at generator side. Do not forget to restore GND if you contunue checkign your circuit.)

Short circuit C18 again and check for IC4 pini 11 to be low.

If you imagine both gates IC4c, d as single AND gate (valve):
At power on it will not propagate the generator signal (at valve input) and keep FETs inactive because pin 9 (valve control) is still not high (5V) (no enable)
As soon pin 9 is high (5V) (valve control enabled) the signal propagates happily the valve and performs its action.

8.
You observed that there are two spare gates being unused up to now. In fact we do not need them. They stay there in parking position up to the time we need them.

Check IC4 pin 3 and pin 6 to be constantly low (0V)

---------------
Now you're done. Congratulations! You discovered the magic of IC4 and mastered it.
Next time we will deal with the powerful heart of the whole PCB - the FET driver. It was thorougly selected in order to enable those FETs for highest performance.
JS

[iankoglin]

Hello all, as you know my problem with the monster board was a jumper to GND, there seems to be two more jumpers to do with feeding the opto, is that correct?.

Have enclosed pics, hope they are clear enough.

Thanks Cornboy.

[IMG][/IMG]


[IMG][/IMG]

G'day Cornboy
The circuit that Sampojo has manufactured was the first that John Stone Drew up where the connections to the Opto were Jumpers over the Gnd he later revised it where he did not need the jumpers as he routed the Gnd land under the opto
Here is the Monster driver T3000 showing the circuit with the Jumpers I think this is the only difference

https://www.dropbox.com/home/John%20...s/BackupFiles#

He also shows double wires/pins ( 2 wires 2 pins one connection) which he explained is to double up on the wires to carry more current or rather to have less resistance As you will notice that the Aux power Gnd and +12v have 2 pins each to carry the double wires, also the 6 pin labelled (GND K A R R X ) only receive 2 double wires/pins (GND,K) (A,R) (R,X) one double wire always goes to (GND.K) the other double wire/pins goes to (R,X) to use the 330 Ohm resistor the or to the (A,R) double wire pins if 330 ohm resistor not requires.
If the 330 ohm resistor is used then the (A,R) pins have to be jumpered I have them permanently soldered together underneath the board

Kindest Regards



Kogs not trying to confuse or perhaps he is

[prochiro]

Bribery will get you everywhere JS

@ JohnStone

WOW, Cornboy is getting the red carpet. I wonder why? Could it be you like Garlic. Anyway, if these notes find there way into Monster V6.0, that will be one grand DIY-PDF. This will also be of extreme help to those who are following. We need all the people we can get to the full build of this project. I also picked up some more understanding of what and why things do what they do. Thanks,
Dana

[Cornboy 555]

Muddling away.

Yes DANA, Garlic is lovely stuff, and i am a dummy, i could have muddled my way through, but this is for everyone, clear and simple, to follow, it's great!!, thanks John Stone.

@ Kogs, thanks for the explanation.

@ John Stone, my function gen is a Victor VC2002, cheapy.
I measured the MA at 5v A/C reads around 90ma DC 10-15ma.
Perhaps You could google it and let me know.

By the time we build our first flying Garlic delivery vehicle, i should have enough for everybody here.

Warm regards Cornboy.

[Midaztouch]

My garlic

@ Cornboy

When you are doing your flying deliverys to the team, just bring my load of garlic to the race track. We will have a BBQ and you can ride the race bike as much as you want!

PS: Kidnap Dana and bring him with you
I have to give the official CREW CHEIF of "TEAM UP" a hug

Keep it Clean and Green
Richie

[Hitby13kw]

John Stone data acquisition unit ~ update

Team, we're almost cooking with gas (at least the stove is lite)
did some more testing with the data acquisition system, have combined the ADS1115 & TMP102 sensors into one script and formatted the output so that each sensor has its own ID tag



A copy of the code can be had for the low low cost of just clicking https://www.dropbox.com/sh/wie0znkfq0agl8o/80SVILHSWf following is the format all data will be sent - if it is preceded with a " #x " this defines what sensor is sending the data.

#1 Temperature sensor with ADD0 to GND - #2Temperature sensor with ADD0 to +3.3V
#3 reserved - - - - - - - - - - - - - - - - - - - - - - - #4reserved
#5 ADC port A0 current sensor - - - - - - - - - #6ADC port A1 voltage sensor
#7 ADC port A2 reserved - - - - - - - - - - - - - #8ADC port A3 reserved
#9 RPM - - - - - - - - - - - - - - - - - - - - - - - - #0reserved




Example of output follows:
#5 873 mV #1 A: 18.19 Celsius #6 3960 mV #2 B: 28.06 Celsius #7 1545 mV #8 3177 mV
#5 870 mV #1 A: 18.12 Celsius #6 3963 mV #2 B: 28.00 Celsius #7 1518 mV #8 3126 mV
#5 858 mV #1 A: 17.87 Celsius #6 3960 mV #2 B: 28.00 Celsius #7 1518 mV #8 3120 mV
#5 4869 mV #1 A: 101.44 Celsius #6 984 mV #2 B: 31.75 Celsius #7 933 mV #8 870 mV
#5 4872 mV #1 A: 101.50 Celsius #6 969 mV #2 B: 31.75 Celsius #7 918 mV #8 855 mV
#5 4788 mV #1 A: 99.75 Celsius #6 969 mV #2 B: 31.69 Celsius #7 918 mV #8 855 mV
#5 4791 mV #1 A: 99.81 Celsius #6 969 mV #2 B: 31.56 Celsius #7 918 mV #8 855 mV
#5 4794 mV #1 A: 99.87 Celsius #6 984 mV #2 B: 31.50 Celsius #7 927 mV #8 855 mV
#5 4788 mV #1 A: 99.75 Celsius #6 984 mV #2 B: 31.50 Celsius #7 930 mV #8 870 mV



@ John, can you work with something like this?

Still pondering how to do the RPM sensor #9 – there are several ways to do this but what I was using caused an interrupt to the Arduino once for every revolution of the motor (we were keying on the falling edge of that signal) this would work, just trying to imagine something better.

Hitby13kw

[iankoglin]

Kogs EBike update

I will try it.
Yesterday I could not wait I connected a normal 50k pot and was running my bike I stalled it and before I could turn it off the 2 1 Ohm resistors on the Monster burned and the machine stopped.
Later I found the Battery is still fully charged I just cannot access any current from it I guess the BMS cut out and I cannot find where to reset it.

I will try the changes you suggest to the Oscillator and replace the Lipo battery with some slab's, fit the twist throttle and see where it goes.
Thanks a million
Please thank your wife for lending you to us.
We all on the these forums really appreciate what you are doing for us.

Kindest Regards to both of you

Kogs

G"Day Team
Thursday I finished building another 2 John Stone's Monsters (untested) also modified my Oscillator to accept the 5k ohms twist throttle. Yesterday I received 3 - 7.1ah SLAB's. Today I finished rehashing my box to fit all the components in and enough room to fit 3 more batteries if I want to. I connected all the components ( including another Motor not connected to my Bike) and connected the batteries every thing went sweet. Tomorrow or Monday I will finish connecting everything permanently and take some pics.

Kindest regards to you all


Kogs still progressing

[Midaztouch]

Twist throttle

@ kogs

Hey kogs,

I have to order a twist throttle from the states. Are you using the Megura twist throttle for your bike? Also, do you have an emergency HV kill button? If so, what is the rating?

Keep it Clean and Green
Richie

[Cornboy 555]

G"Day Team
Thursday I finished building another 2 John Stone's Monsters (untested) also modified my Oscillator to accept the 5k ohms twist throttle. Yesterday I received 3 - 7.1ah SLAB's. Today I finished rehashing my box to fit all the components in and enough room to fit 3 more batteries if I want to. I connected all the components ( including another Motor not connected to my Bike) and connected the batteries every thing went sweet. Tomorrow or Monday I will finish connecting everything permanently and take some pics.

Kindest regards to you all


Kogs still progressing

Great Kogs, look forward to it.

Regards Cornboy.

[Cornboy 555]

Jump Again.

Hello John Stone and all, Finished Instructable Section 3 Opto input.

Works beautiful, but one catch, you must use the jumpers shown on the board, you can connect them underneath if you wish.

Let's go with next step, when you'r ready John.

Thanks Heaps, Cornboy.