I'm not sure why there are two, but one diode can be useful when in
circuit programming. It prevents the high voltage that you need to
apply to MCLR from feeding back into the +5V line, possibly damaging
the PIC, other components, and the programmer itself.
Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
-Douglas Adams
> I have seen some schematics with a PIC included which have had two
> diodes on the MCLR pin connected
> like my mighty fine ;-) ASCII schema below.
>
> VCC
> |
> |
> | | 10K
> | |
> |
> ------------
> | |
> Diod V V Diod
> | |
> | |
> _____________________ MCLR
>
> What function does the diodes have?
Looks like it's there for isolating the Vcc line from the MCLR line during
ICSP. Not sure why two though, but might guess that it has something to do
with the reverse current limit?
There is another arrangement when the diode goes the other way (pointing
towards Vcc), and there is a capacitor between the MCLR line and ground as
well -- it's an RC startup delay, with the diode helping the capacitor
discharge quickly after power off.
Cheers,
-Neil.
On Tuesday 19 April 2005 03:09 pm, Ake Hedman scribbled: {Quote hidden}
> I have seen some schematics with a PIC included which have had two
> diodes on the MCLR pin connected
> like my mighty fine ;-) ASCII schema below.
>
> VCC
>
> | | 10K
>
> ------------
>
> Diod V V Diod
>
>
> _____________________ MCLR
>
> What function does the diodes have?
>
> /Ake
In this schematic, not only they have nonsens, but also they can do an
inexpected behaviour of the PIC, because the current flowing into the
MCLR could be too small if VCC is 3.3V and/or R is greater than 10K.
There is no need of any diode connected like this when programming a
PIC, using HVP and large enough pull up resitor to Vcc from MCLR.
> I have seen some schematics with a PIC included which have had two
> diodes on the MCLR pin connected
> like my mighty fine ;-) ASCII schema below.
>
> VCC
> |
> |
> | | 10K
> | |
> |
> ------------
> | |
> Diod V V Diod
> | |
> | |
> _____________________ MCLR
>
> What function does the diodes have?
>
> /Ake
>
> --
> ---
> Ake Hedman (YAP - Yet Another Programmer)
> eurosource, Brattbergavägen 17, 820 50 LOS, Sweden
> Phone: (46) 657 413430 Cellular: (46) 73 84 84 102
> Company home: http://www.eurosource.se
> Kryddor/Te/Kaffe: http://www.brattberg.com
> Personal homepage: http://www.eurosource.se/akhe
> Automated home: http://www.vscp.org
>
This is also the recommendation form Microchip. No diodes, no capacitors just a 10K resistor to VDD.
I have also seen a transistor used instead of the diodes. And NPN with the base connected directly connected to the emitter which goes to MCLR and the collector to a resistor which is connected to VDD. But I guess this is a diode.
So two diodes is just double nonsense then...? ;-)
/Ake
>In this schematic, not only they have nonsens, but also they can do an
>inexpected behaviour of the PIC, because the current flowing into the
>MCLR could be too small if VCC is 3.3V and/or R is greater than 10K.
>There is no need of any diode connected like this when programming a
>PIC, using HVP and large enough pull up resitor to Vcc from MCLR.
>
>Vasile
>http://surducan.netfirms.com
>
>
>On 4/19/05, Ake Hedman <akheKILLspameurosource.se> wrote:
>
>
>>I have seen some schematics with a PIC included which have had two
>>diodes on the MCLR pin connected
>>like my mighty fine ;-) ASCII schema below.
>>
>> VCC
>> |
>> |
>> | | 10K
>> | |
>> |
>> ------------
>> | |
>> Diod V V Diod
>> | |
>> | |
>> _____________________ MCLR
>>
>>What function does the diodes have?
>>
>>/Ake
>>
>>--
>>---
>>Ake Hedman (YAP - Yet Another Programmer)
>>eurosource, Brattbergavägen 17, 820 50 LOS, Sweden
>>Phone: (46) 657 413430 Cellular: (46) 73 84 84 102
>>Company home: http://www.eurosource.se
>>Kryddor/Te/Kaffe: http://www.brattberg.com
>>Personal homepage: http://www.eurosource.se/akhe
>>Automated home: http://www.vscp.org
>>
> This is also the recommendation form Microchip. No diodes, no capacitors
> just a 10K resistor to VDD.
The 10K to Vdd should work just fine, unless the Vdd power supply isn't
'stiff' enough to avoid rising in voltage with the injected (Vpp-Vdd)/10k
current. This is highly unlikely to be an issue except in the very lowest
power designs.
>> This is also the recommendation form Microchip. No diodes, no
>> capacitors just a 10K resistor to VDD.
>
>
> The 10K to Vdd should work just fine, unless the Vdd power supply
> isn't 'stiff' enough to avoid rising in voltage with the injected
> (Vpp-Vdd)/10k current. This is highly unlikely to be an issue except
> in the very lowest power designs.
>
> Bob Ammerman
> RAm Systems
>
On 4/19/05, Ake Hedman <.....akheKILLspam.....eurosource.se> wrote:
> I have seen some schematics with a PIC included which have had two
> diodes on the MCLR pin connected
> like my mighty fine ;-) ASCII schema below.
>
> VCC
> |
> |
> | | 10K
> | |
> |
> ------------
> | |
> Diod V V Diod
> | |
> | |
> _____________________ MCLR
>
> What function does the diodes have?
While others have commented on the function of having a diode in that
position, there may be two diodes on the schematic because they don't
know which diode they are going to use. They will only populate a
single diode position when the device is built. Seems unusual with a
diode, but perhaps the board design is made for two different assembly
houses with two cheapest diodes whose land patterns don't overlap very
well. If you have the board layout to look at, you might even find
that one is a diode and the other is actually a transistor used in a
diode configuration.
>On 4/19/05, Ake Hedman <EraseMEakhespam_OUTTakeThisOuTeurosource.se> wrote:
>
>
>>I have seen some schematics with a PIC included which have had two
>>diodes on the MCLR pin connected
>>like my mighty fine ;-) ASCII schema below.
>>
>> VCC
>> |
>> |
>> | | 10K
>> | |
>> |
>> ------------
>> | |
>> Diod V V Diod
>> | |
>> | |
>> _____________________ MCLR
>>
>>What function does the diodes have?
>>
>>
>
>While others have commented on the function of having a diode in that
>position, there may be two diodes on the schematic because they don't
>know which diode they are going to use. They will only populate a
>single diode position when the device is built. Seems unusual with a
>diode, but perhaps the board design is made for two different assembly
>houses with two cheapest diodes whose land patterns don't overlap very
>well. If you have the board layout to look at, you might even find
>that one is a diode and the other is actually a transistor used in a
>diode configuration.
>
>Bradley
>
>
>
At 21.08 2005.04.20 +0200, you wrote:
>And the winner is.... Bradley Ferguson!!
>
>You are absolutely right. There is actually a transistor mounted on the board. Mystery solved.
BTW: exactly in what regard does a transistor connected that way behave better than a diode?
>Thanks all for enlighten me!
>
>Cheers
>/Ake
>
>Bradley Ferguson wrote:
>
>>On 4/19/05, Ake Hedman <akhespam_OUTeurosource.se> wrote:
>>
>>
>>>I have seen some schematics with a PIC included which have had two
>>>diodes on the MCLR pin connected
>>>like my mighty fine ;-) ASCII schema below.
>>>
>>> VCC
>>> |
>>> |
>>> | | 10K
>>> | |
>>> |
>>> ------------
>>> | |
>>> Diod V V Diod
>>> | |
>>> | |
>>> _____________________ MCLR
>>>
>>>What function does the diodes have?
>>>
>>
>>While others have commented on the function of having a diode in that
>>position, there may be two diodes on the schematic because they don't
>>know which diode they are going to use. They will only populate a
>>single diode position when the device is built. Seems unusual with a
>>diode, but perhaps the board design is made for two different assembly
>>houses with two cheapest diodes whose land patterns don't overlap very
>>well. If you have the board layout to look at, you might even find
>>that one is a diode and the other is actually a transistor used in a
>>diode configuration.
>>
>>Bradley
>>
>>
>
>
>--
>---
>Ake Hedman (YAP - Yet Another Programmer)
>eurosource, Brattbergavägen 17, 820 50 LOS, Sweden
>Phone: (46) 657 413430 Cellular: (46) 73 84 84 102
>Company home: http://www.eurosource.se
>Kryddor/Te/Kaffe: http://www.brattberg.com
>Personal homepage: http://www.eurosource.se/akhe
>Automated home: http://www.vscp.org
>
>
>At 21.08 2005.04.20 +0200, you wrote:
>
>
>>And the winner is.... Bradley Ferguson!!
>>
>>You are absolutely right. There is actually a transistor mounted on the board. Mystery solved.
>>
>>
>
>BTW: exactly in what regard does a transistor connected that way behave better than a diode?
>
>
>
>
Isn't it just that they got the transistor at a lower cost then the diode?
>
>
>
>>Thanks all for enlighten me!
>>
>>Cheers
>>/Ake
>>
>>Bradley Ferguson wrote:
>>
>>
>>
>>>On 4/19/05, Ake Hedman <@spam@akheKILLspameurosource.se> wrote:
>>>
>>>
>>>
>>>
>>>>I have seen some schematics with a PIC included which have had two
>>>>diodes on the MCLR pin connected
>>>>like my mighty fine ;-) ASCII schema below.
>>>>
>>>> VCC
>>>> |
>>>> |
>>>> | | 10K
>>>> | |
>>>> |
>>>> ------------
>>>> | |
>>>>Diod V V Diod
>>>> | |
>>>> | |
>>>> _____________________ MCLR
>>>>
>>>>What function does the diodes have?
>>>>
>>>>
>>>>
>>>While others have commented on the function of having a diode in that
>>>position, there may be two diodes on the schematic because they don't
>>>know which diode they are going to use. They will only populate a
>>>single diode position when the device is built. Seems unusual with a
>>>diode, but perhaps the board design is made for two different assembly
>>>houses with two cheapest diodes whose land patterns don't overlap very
>>>well. If you have the board layout to look at, you might even find
>>>that one is a diode and the other is actually a transistor used in a
>>>diode configuration.
>>>
>>>Bradley
>>>
>>>
>>>
>>>
>>--
>>---
>>Ake Hedman (YAP - Yet Another Programmer)
>>eurosource, Brattbergavägen 17, 820 50 LOS, Sweden
>>Phone: (46) 657 413430 Cellular: (46) 73 84 84 102
>>Company home: http://www.eurosource.se
>>Kryddor/Te/Kaffe: http://www.brattberg.com
>>Personal homepage: http://www.eurosource.se/akhe
>>Automated home: http://www.vscp.org
>>
>>
On 4/21/05, ThePicMan <KILLspamthepicmanKILLspaminfinito.it> wrote:
> At 21.08 2005.04.20 +0200, you wrote:
> >And the winner is.... Bradley Ferguson!!
> >
> >You are absolutely right. There is actually a transistor mounted on the board. Mystery solved.
>
> BTW: exactly in what regard does a transistor connected that way behave better than a diode?
I believe other than cheaper cost, fewer parts/board, higher
part-volume, if you use the C-B junction as the diode you have the
entire reverse breakdown voltage of the transistor available to you.
This breakdown voltage might be higher than the standard "jelly bean"
diode used by the assembler.
> At 21.08 2005.04.20 +0200, you wrote:
>> And the winner is.... Bradley Ferguson!!
>>
>> You are absolutely right. There is actually a transistor mounted on
>> the board. Mystery solved.
>
> BTW: exactly in what regard does a transistor connected that way
> behave better than a diode?
In the way that it can be pushed into (undesirable) reverse breakdown of
the BE junction under certain conditions. This occurs at -4..-9V Vbe.
> I believe other than cheaper cost, fewer parts/board, higher
> part-volume, if you use the C-B junction as the diode you have the
> entire reverse breakdown voltage of the transistor available to you.
> This breakdown voltage might be higher than the standard "jelly bean"
> diode used by the assembler.
I do not agree. A jellybean diode will have Vr = 80V (1N4148).
Transistors with such specs are not 'jellybean'. The only time it makes
sense to use a transistor as a diode is when you already use
transistors in the circuit, and no diodes. Then the stocking problems
are reduced and the equivalent price of the transistor is much lower
since it is bought in bulk.
>On Thu, 21 Apr 2005, ThePicMan wrote:
>
>>At 21.08 2005.04.20 +0200, you wrote:
>>>And the winner is.... Bradley Ferguson!!
>>>
>>>You are absolutely right. There is actually a transistor mounted on the board. Mystery solved.
>>
>>BTW: exactly in what regard does a transistor connected that way behave better than a diode?
>
>In the way that it can be pushed into (undesirable) reverse breakdown of the BE junction under certain conditions. This occurs at -4..-9V Vbe.