[Uksa007]

So I have received my boards and assembled one.

Preliminary testing at the moment and it’s working great.
The Boost circuit produces about 22.5v.

My software modifications are working great, see fork link above.

Tested with some solenoids and they open and stay open OK with no water pressure.
However when there is water pressure they open OK, but then close almost immediately (after the booster is turned off).
Seems the 5V is not sufficient to keep them open.
I’m using the below 12V DC solenoids
http://www.ebay.com.au/itm/3-4-DC12V-PP-N-C-Solenoid-Electric-Valve-Water-Control-Diverter-Device-New/381978081748

Ray does the OpenSprinker DC drive the solenoids OK when run from 5V eg USB?

Solved my issues:
There was too much resistance in the cable from the OSPi Mini to Solenoids (10m run), resulting inadequate current to hold it open.
Fix: Doubled the conductors and she working like it should!

Attachments:

1. 

   OSPI-mini-crop.jpg
   

[Uksa007]

My PCB’s and parts are on order.

While I’m waiting on hardware I would like to improve the firmware “hacks” I made.
I have forked the Opensprinkler-Firmware. https://github.com/Uksa007/OpenSprinkler-Firmware
Thinking I should probably create a new hardware type for my OSPi-DC, as it is fairly different, no RTC, Buttons, or ADC and as a result the pin defines are a little different.

defines.h seems to define the hardware and pins, whats the best way to go about creating a different hardware version of the OSPI?

/** Hardware version base numbers */
#define OSPI_HW_VERSION_BASE 0x40

/** OSPi pin defines */
#if defined(OSPI)

Have gone with creating a new hardware type OSPIMINI and have updated many files to include new hardware type.

[Uksa007]

Hi Ray,

Thanks for taking the time to look at the substitute parts.

3) Q1-8 the N-MOSFETs should all be AO3400,
For some reason I had it in my mind that the 74HC595 latch was using 5V Vcc, but you are correct it using 3V3, so the N-Mosfet will be limited to 1A which might be a bit low, not really sure of the inrush current required to turn on the average solenoid.

I too was surprised that the AO3400 is not available from Element14
This seems to be Element14 substitute so I might go with it
http://www.farnell.com/datasheets/1911844.pdf?_ga=2.122560779.1206234277.1497427852-1758119718.1492662898

4) Booster circuit optimized for 5V input.
I have been scratching my head trying to work out how you came up with the component values.
OSDC uses RT 27K and RB 1.6K, CT 330pF
OSBEE Uses RT 75K and RB 4.7K, CT 470pF

Using this calculator I get close to your values for OSDC, using a 9V input (OSDC transformer)
I guess I will go with these and can adjust if need be.
http://www.gmsystems.com/switching-reg-calculator-for-mc-34063-or-mc33063.html

Using these values for input
Vin 9V
Vout 22V
Iout 55 mA
Vripple 1.8 mV(pp)
Fmin 77kHz

It calculates the below, of note are the R1 and R2 which are the voltage divider for the comparator are way off from your values for OSBee R1 and R2?
Ct=325 pF
Ipk=294 mA
Rsc=1.02 Ohm
Lmin=221 uH
Co=2236 uF
R=180 Ohm
R1=1.2k R2=20k (22.08V)

Thanks again.

[Uksa007]

Thanks Ray,

I have been hard at work and have created PCB layout about to send it off to be fabricated.

Couple of questions re substitute parts:
1) The 220uH inductor for the boost circuit, are there any specs for current and resistance?
One available locally is RLB0912-221KL 220uH, RMS current 440 mA, 0.74 ohms is this one suitable?
http://www.farnell.com/datasheets/1647944.pdf?_ga=2.80422039.1263943725.1497055596-1758119718.1492662898

2)QP1, QP2 AP3401 or BSS84
The P-Channel FETs to turn on the boost and fire it.
AP3401 doesn’t seem to exists(miss-type?), should this be AO3401?
I assume that the BSS84 can’t handle enough current?
AO3401 isn’t available locally will a DMG2307L be ok, datasheet below?
http://www.farnell.com/datasheets/1915926.pdf?_ga=2.41980133.111771980.1497254896-1758119718.1492662898

3) N-Channel FETs to switch the Solenoids to GND, Q1-8 AO3400 or BSS123
AO3400 not available locally, assume BSS123 can’t handle enough current?
ZXMN3F30FHTA is available locally will it be ok, data sheet below?
http://www.farnell.com/datasheets/2097584.pdf?_ga=2.41015650.111771980.1497254896-1758119718.1492662898

4)Should I optimize the Boot circuit to operate more efficiently on 5V input voltage?
I notice you different values in the OSBee for the capacitor and resistor around the 34063 boost converter.
Is the Osbee optimized for 5V input?

Thanks for your time!

[Uksa007]

Answering my own questions, I found the pin assignments in define.h and modified the Opensprinker.cpp to define the hardware as DC with no current sensing.
Rebuilt Opensprinkler and now my About page reports:

Hardware: OSPi – DC

YAY I will take that as a WIN!!!!!

It’s a bit of a hack, but for testing it’s a start!

[Uksa007]

So got the unified firmware running on the Pi, have not found any options to configure the hardware.

Its reporting:
Hardware OSPi – AC

I probably need to create a new hardware type or modify one of the existing hardware types.
Think I’m going to need a OSPi mini – DC, eg OSPi with DC support.
Is there any way to configure the hardware types using configuration files, or do I need to crack open the source?

anyone?

[Uksa007]

I have made some progress and made some executive decisions, which can easily be changed:

1) Opted to use the Single FET design, makes more sense to me.
1) BST_PWR (which powers up the boost circuit) has been connected to GPIO1
2) BST_E (which turns on the +24V to the COM terminals) has been connected to GPIO4

Other things I have tried to keep as standard from the original OSPi eg:
Data D5 = 13-GPIO2
Clock D6 = 7-GPIO7
Latch D7 = 15-GPIO3
!OE A1 = 11-GPIO0

ADC:
So I was trying to avoiding having to include an ADC.
How important is it that the SENSE voltage is available to the software?
Is it used for information purposes or protection to ensure you are not driving too much current?
Can I omit it?

Attachments:

1. 

   Uksa007-OSPi.jpg
   

[Author]

Posts