- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi every body!
I have a CY8CKIT - 042 PSoC® 4 Pioneer Kit and 6 P10 LED Matrix Panels 16x32 (1 colour) connected in series. I'm using PSoC Creator 3.0 SP2.
Link about P10 LED Matrix Panels :
http://forum.arduino.cc/index.php?topic=260320.0
http://arectron.com/yazilim/p10-panel-kullanimi.html
I want change brightness of P10 LED Matrix Panels via OE pin.
Example: P2.0 on CY8CKIT - 042 PSoC® 4 Pioneer Kit connected to OE pin on the first P10 LED Matrix Panels.
Beside using PWM component, can you please tell me about another solutions to change brightness?
Many thanks!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Use the PrISM module ?
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Dana,
I'll try to code for PrISM.
Have Voltage Digital to Analog Converter component in PSoC Creator?
Can I use it to change brightness?
Many thanks!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
i don´t have any question about this topic but about deadtime.
i have problem producing dead band. I have read some of your contribution on that subject, and would be happy if you can help me.
My code is working very well apart of the issue with positve bandwith of the two phase outputs signals, that are differnts. Normally they should be equal.
Do you have any idea, what the problem is?
Thanks for repying
Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Have Voltage Digital to Analog Converter component in PSoC Creator?
Can I use it to change brightness?
In your case the enable signal on the P10 is a digital signal, so the IDAC8's
on PSOC cannot be used for this purpose. Note its an IDAC and you create a VDAC
out of it by using a TIA or a resistor.
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks for your supports!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
My code is working very well apart of the issue with positve bandwith of the two phase outputs signals, that are differnts. Normally they should be equal.
Do you have any idea, what the problem is?
The way 2 phase signals work is they are replicas 180 degrees out of phase. Same
frequency and duty cycle. But that also means they would have no dead time. In order
to create deadtime, still keeping the frequency the same, then one pulse's duty cycle must
be lower than the other. So when you set up deadtime in the PWM configuration, you see
phase 2 has a lower duty cycle by 2 X the number of deadtime clocks.
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello dana,
thank you so much for your support. I understand now why the positives pulses are differents from each other.
Please could you help me to produce deaddand while keeping the bandwidth equal? The Frequencys can be different.
Thanks a lot again
Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Dear Rosine
To help you and others with a similar problem it would be a lot better to start a new thread for your question with a meaningful description instead of hijacking a thread with a quite different theme.
When later someone searches for PWM-threads he will find yours (and others) as expected.
Bob
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Rosine, can you describe a little more why you need exactly same
PW on both phases ? What you are driving downstream that is driving
this specification ?
Also what pulse width range do you need, and what frequency of pulse
train is needed.
In general Verilog solution would be one way to achieve this. Or use a
LUT to trigger 2 PWM channels operating in one shot mode, that would
sequence the pulses, and with a combinationj of compare and period
writes (using PWM APIs) you could manage deadtime and frequency.
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Did you try the TCPWM component in pwm mode with deadtime insertion and right align?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Eric that component creates unequal duty cycle outputs to achieve deadtime,
which poster does not want.
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
Thank all of you for replying.
With the code, i would like to control a brodge made of MOSFETs. From the requirement i received, the Pulswidth, Dead time must be equal.
your right but by using two PWMs, after flashing the code in the controler, i am just geting one output signal right, on the other output, there is no signal.
Please can you give an example, which is working?
Thank again very much
Regards
rosine
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Rosine, post your project so we can take a look at it.
“File” Creator
“Create Workspace Bundle”
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
on a subject "PWM Phase Shift", you have propose a method:
" Another way is to use 2 PWMs. Start one, then using a counter count the first PWMs clock and start the other after a desired number of clock delays. Use HW enables to do this. Both PWMs would have the same period, compare values."
Here is the link to your suggestion:
http://www.cypress.com/?app=forum&id=4749&rID=102369
That was your suggestion, but how can you realise this? Please I think it is what i need. I have try to do it but i am getting only one output properly.
I will be please with any help
Regards
Rosine
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The method I suggested would eat up some resources on PSOC 4.
Let me ask another question, what is the frequency you want to operate
PWMs at, the range of duty cycle, and the resolution of duty cycle control.
Reason I ask is the approach that shortens one PWM out duty cycle to
create the deadtime would be a small fraction difference over most of the
duty cycle range. So if you are led driving that effect would be small,
as the eye tends to integrate brighness. And if you are working with
leds the deatime you need probably ony 1 - 2 clocks, so in a 16 bit PWM
that difefrence in brightness is lost in the "noise".
Try it in a test bed, that should take < 5 min to set up, thats the beauty of PSOC,
how fast you can do quick tests and checks of ideas.
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Here is an example, deadband is 2 clocks on each side of
PWM2 out. So thats 4 clocks out of 500 or .8% of the duty cycle.
Below is drawn to scale. The 2 clocks is ~ 200 nS of deadtime.
I drove the PWM with a 10 Mhz clock. You can increase DT and
period to affect other criteria. Like if you wanted 1 uS of DT and still
keep delta < 1 %. In your application I do not think you will be able to
observe any "artifacts" due to small differences in Duty Cycle between
outputs.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Here is an example, deadband is 2 clocks on each side of
PWM2 out. Clk to PWM 24 Mhz. So thats 4 clocks out of 1199
or .3% of the duty cycle. Below is drawn to scale. The 2 clocks
is ~ 80 nS of deadtime. You can increase DT and period to affect
other criteria.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
i would like to know how to calculate the switching frequency from the Turn-on delay time [td(on)], Rise time(tr), Turn-off delay time (td(off)) and Fall time(tf).
Thank for replying
Regards
Rosine
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The PWM out switching freq = Fpwmclkin / Period, period is the
setting you invoke in its properties or by using API to write period.
The turn on delay or in this case deadtime is the number of input
clks you set in properties, or again by API write. The dead ime is
= DTclks / Fpwmclkin. So if your Fpwmclkin = 1 Mhz (1 uS period)
and DTclks = 2, then deadtime = 2 uS.
Rise and fall times speced for GPIO in datasheet.
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
the PWMs will be produce to drive the Full-Bridge. I was thinking, that the switching frequency of the MOSFETs is very important
in order to calculate the accurated dead time.
Thanks
Regards
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Its preferred that you use an integrated H Bridge as PSOC
outputs cannot create fast MOSFET gate drive as the GPIO
is very limited in source/sink current. The integrated H Bridge
will have timing specs so you create a dead time value to handle
worst case potential overlap in the H Bridge source and sink MOSFET
swicthing.
On another matter, if you are switching large inductive currents
board layout very important, especially Power and Ground. See
attached.
http://www.cypress.com/?rID=39677 AN57821 - PSoC® 3, PSoC 4, and PSoC 5LP Mixed Signal Circuit Board Layout Considerations
http://www.cypress.com/?rID=40247 AN58827 - PSoC® 3 and PSoC 5LP Internal Analog Routing Considerations
http://www.ti.com/general/docs/lit/getliterature.tsp?literatureNumber=sdya011&fileType=pdf
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
is there an asynchronous Toggle Flip Flop by cypress?
Thank for the reply.
Regards
Rosine
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
See http://www.cypress.com/?rID=73664
How 'asynchronous' does it need to be - a toggle FF always has a clock input...
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This might also be of use -
http://www.cypress.com/?rID=80799 Using clocks as data PSOC 4
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Dana,
can you give me an example how to create a software interrupt?
Thank you so much for replying
Regards
Claire
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Claire, tell me what you are trying to accomplish -
1) What will you use to trigger the ISR
2) What the ISR has to do
3) Is it as one time only ISR or a periodic one ?
4) Why is it software only ?
Regards, Dana.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content