View previous topic :: View next topic |
Author |
Message |
eaton
Joined: 02 Oct 2004 Posts: 22 Location: cleveland
|
reset supervisor |
Posted: Fri Jan 06, 2006 5:58 am |
|
|
Has anyone used a MCP111 voltage supervisor? I need to maintain the ICSP capability of my chip (18F252) but I don't understand the circuit shown in the data sheet. It shows the connection to pin 1(MCLR) through a pullup R to Vpp/Vdd. I don't get the Vdd part. Why have the pullup if the MCP111 is providing the reset? If the connection labeled Vpp/Vdd is truly connected to Vdd then it would seem that during programming the 5V supply for the entire circuit would raised to Vpp.
As always any help is greatly appreciated. |
|
|
Ttelmah Guest
|
|
Posted: Fri Jan 06, 2006 6:41 am |
|
|
You raise the 'point' Vdd/Vpp to the programming voltage. The 1K resistor from the MCLR pin to the supervisor chip, limits the current flowing, but the chip is rated to stand the programming voltage on this pin (this is why only one version is suitable). Have you pulled down TB087?. This gives tables for the resistor values needed, and which supervisor chips are suitable for particular PIC programming voltages. Note particularly figure 5, which shows how to avoid the tendency to raise the entire rail, when ICSP is used.
Best Wishes |
|
|
Humberto
Joined: 08 Sep 2003 Posts: 1215 Location: Buenos Aires, La Reina del Plata
|
|
Posted: Fri Jan 06, 2006 6:41 am |
|
|
Quote: |
I need to maintain the ICSP capability of my chip (18F252)
|
You get it. The spec of this device allows to be used with the ICSP.
See NOTE 3 in Page 3.
Quote: |
but I don't understand the circuit shown in the data sheet.It shows the connection to pin 1(MCLR) through a pullup R to Vpp/Vdd.
|
If you are using ICSP, already you have the MCLR circuit wired with a pull up resistor (47K) so it will not be necesary to add a pull up resistor. Anyway to protect the MCP111 output of rising voltage (while programming secuence) Microchip recomend a 1K resistor connected as in Figure 4.4.
Quote: |
I don't get the Vdd part
|
You get it, Operating Voltage Range: 1.0V to 5.5V. (P3)
Quote: |
Why have the pullup if the MCP111 is providing the reset?
|
The output of the MCP111 is an open drain type. If VDD voltage falls below of the voltage trip point, the output of the MCP111 pull down the MCLR pin of the PIC causing a RESET.
Quote: |
If the connection labeled Vpp/Vdd is truly connected to Vdd then it would seem that during programming the 5V supply for the entire circuit would raised to Vpp.
|
No. The Vpp/Vdd doesn't rise while programming, only the MCLR and Vout.
Humberto |
|
|
Ttelmah Guest
|
|
Posted: Fri Jan 06, 2006 6:43 am |
|
|
Er. Humberto. Vpp, _is_ the voltage on MCLR. It is the programming voltage, and it does raise to program the chip...
Best Wishes |
|
|
Humberto
Joined: 08 Sep 2003 Posts: 1215 Location: Buenos Aires, La Reina del Plata
|
|
Posted: Fri Jan 06, 2006 6:50 am |
|
|
RJ you are right.
Vpp is the programming voltage.
Vdd is the operating voltage.
But in Figure 4.4 there are is a mistake showing them as Vdd/Vpp. I gues this is the cause of the of the confusion of the poster.
Should be Vpp only.
regards,
Humberto |
|
|
Ttelmah Guest
|
|
Posted: Fri Jan 06, 2006 6:59 am |
|
|
Yes.
However that is because on the data sheets for the chip itself, the pin is labelled as Vdd/Vpp. This voltage 'point', has to either be at Vpp for programming, or at Vdd for running. This is why they labelled it this way, but they then didn't explain that circuitry will be necessary to make the pin go to the two different voltages, without crosslinking the supply. Either just a simple 'jumper', or diode isolation...
It is 'fun'. :-)
The Tech note, makes it much clearer.
Best Wishes |
|
|
Humberto
Joined: 08 Sep 2003 Posts: 1215 Location: Buenos Aires, La Reina del Plata
|
|
Posted: Fri Jan 06, 2006 7:19 am |
|
|
Quote: |
However that is because on the data sheets for the chip itself, the pin is labelled as Vdd/Vpp.
|
RJ I'm not agree.
All the references to the operating voltages in the datasheet are named Vdd. Except in Figure 4.4
Vdd for any reason will rise. Only the Vout wich is an open drain transistor connected to MCLR.
Vpp is the programming voltage and is applicable only as a PIC feature, it is not a MCPxxx pin and is not caracterized.
regards,
Humberto |
|
|
Charlie U
Joined: 09 Sep 2003 Posts: 183 Location: Somewhere under water in the Great Lakes
|
|
Posted: Fri Jan 06, 2006 7:27 am |
|
|
The overall concept is descibed in Microchip application note AN820. I have been using this concept for a couple of years with excellent results. See if this app. note helps.
Charlie |
|
|
eaton
Joined: 02 Oct 2004 Posts: 22 Location: cleveland
|
|
Posted: Fri Jan 06, 2006 6:47 pm |
|
|
thank you all,
I think the diode isolation shown in fig 5 of TB087 is the way to go. |
|
|
kender
Joined: 09 Aug 2004 Posts: 768 Location: Silicon Valley
|
|
Posted: Fri Jan 06, 2006 10:42 pm |
|
|
I have successfully used the topology shown in the Fig.2 of the TB087
Rs = 13k
Rpu = 47k
Vcc = +5V
EDIT: The PIC was programmed in high voltage programming mode.
Last edited by kender on Sat Jan 07, 2006 3:51 am; edited 1 time in total |
|
|
Ttelmah Guest
|
|
Posted: Sat Jan 07, 2006 3:24 am |
|
|
Humberto wrote: | Quote: |
However that is because on the data sheets for the chip itself, the pin is labelled as Vdd/Vpp.
|
RJ I'm not agree.
All the references to the operating voltages in the datasheet are named Vdd. Except in Figure 4.4
Vdd for any reason will rise. Only the Vout wich is an open drain transistor connected to MCLR.
Vpp is the programming voltage and is applicable only as a PIC feature, it is not a MCPxxx pin and is not caracterized.
regards,
Humberto |
Look at the pin diagram for the chip. Look at the label on the MCLR pin.
Also look at the 'memory programming requirements' section of the DC characteristics.
The programming application note also covers this. MCLR can have three states. Low for reset, Vdd for normal operation, and Vpp for programming. ICSP, with a voltage monitor circuit, involves allowing all three states to be achieved, without disconnecting the chip from the board.
Best Wishes |
|
|
libor
Joined: 14 Dec 2004 Posts: 288 Location: Hungary
|
|
Posted: Sat Jan 07, 2006 4:26 am |
|
|
Sorry guys for the slightly off-topic question in this thread.
I have not yet used reset supervisors in my applications. Should I? Is it really necessary to include the supervisor chip to a PIC that has its own in-built power-up-timer, brown-out-detector, low-voltage-detector, watchdog etc.
Is there a situation when only a separate supervisor can ensure your application's power-up/down stability ? |
|
|
eaton
Joined: 02 Oct 2004 Posts: 22 Location: cleveland
|
|
Posted: Sat Jan 07, 2006 7:17 am |
|
|
I'm still unclear on the connection of Rpu in fig.2 of TB087. if Vpp is applied here during programming and this resistor is 'normally' connected to Vdd then Vdd will rise to the Vpp level during programming. Won't this affect the rest of the ckt unless something is there (diode) to provide isolation during programming? |
|
|
Humberto
Joined: 08 Sep 2003 Posts: 1215 Location: Buenos Aires, La Reina del Plata
|
|
Posted: Sat Jan 07, 2006 9:45 am |
|
|
Ttelmah wrote:
Quote: |
You raise the 'point' Vdd/Vpp to the programming voltage.
|
And I think is the point in this discussion.
My position is that Vdd/Vpp is a 'typo' error in the Microchip datasheet and Technical Bulletins.
Vdd, Vout, Vpp, Vss, Gnd etc are used universally and each one have an univoque meaning.
Hence:
Vpp = is the programming voltage.
Vdd = is the operating voltage.
By no way Vdd/Vpp are equivalents nor could be wired toghether unless Vpp be equal to Vdd.
Ttelmah wrote:
Quote: |
Look at the pin diagram for the chip.
|
Page 1 of the datasheet:
PIN1 Vout
PIN2 Vss
PIN3 Vdd
Being this valid for all the packages: SOT23 SOT-89 3/SC-70 TO-92
There is not any reference nor PIN names different than Vdd, Vss Vout.
Ttelmah wrote:
Quote: |
Look at the label on the MCLR pin. Also look at the 'memory programming requirements' section of the DC characteristics.
The programming application note also covers this.
Also look at the 'memory programming requirements' section of the DC characteristics. The programming application note also covers this.
|
All this references point to places refering to PIC microcontrollers characteristic, not to the MCP111.
Ttelmah wrote:
Quote: |
MCLR can have three states. Low for reset, Vdd for normal operation, and Vpp for programming.
|
This is the MCLR function in any PIC microcontroller.
eaton wrote:
Quote: |
I'm still unclear on the connection of Rpu in fig.2 of TB087. if Vpp is applied here during programming and this resistor is 'normally' connected to Vdd then Vdd will rise to the Vpp level during programming.
|
Of course. For the same reason I explained. Vdd should not rise for any reason ! ! !
To put this discussion black over white, @eaton you have 2 choices:
1) Follow the procedure I suggest, that is: ignore the label Vdd/Vpp and connect
at this point ONLY the Vpp voltage wired to MCLR.With the 1K resistor protection.
2) Follow the datasheet procedure, wiring Vdd and Vpp toghether
and surelly you will burn all the silicon aroud.
Now you play...
Keep well,
Humberto
Last edited by Humberto on Sat Jan 07, 2006 9:53 am; edited 1 time in total |
|
|
Ttelmah Guest
|
|
Posted: Sat Jan 07, 2006 9:52 am |
|
|
They don't claim they are 'equivalents'.
The point is that you attach the Vpp/Vdd voltage to this point interchangeably, to this point for ICSP. The supply supervisor, supports having it's output pin taken up to Vpp, without damage. They are simply, and correctly, showing one way of connecting the system for ICSP. It requires that the pullup resistor is low enough, so that the droop still allows programming, but with modern chips, the programming current is so low, that this is not a problem.
The table in the manual for the supply monitor circuit, gives the resistor values needed
Hence the connection point, becomes the one for Vpp/Vdd.
Best Wishes |
|
|
|