Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

ICD2 USB clone - how to get it to work with 3.3V PIC24s ???

Status
Not open for further replies.
3v3 pic

polymath said:
Hi szlovak
I do not think Alan69 has tried his own suggestion.



I use it all the time, but of course on normal inputs not schmitt trigger.


I don't use PIC24, but assuming it's inputs are schmitt trigger too, I would try dropping ICD to 4.1V and inserting just series resistors into the lines. 4*.8 = 3.28, and backwards 5*.8 = 4, so you should be able to talk to both 3.3V and 5V targets without changing, let the resistors and input diodes take care of the rest. Have the ICD and targets right in front of me, I'll try it in just a little and make sure it's good, could be handy and almost no extra parts..
 

2v level shift

Alan69 said:
polymath said:
Hi szlovak
I do not think Alan69 has tried his own suggestion.



I use it all the time, but of course on normal inputs not schmitt trigger.


I don't use PIC24, but assuming it's inputs are schmitt trigger too, I would try dropping ICD to 4.1V and inserting just series resistors into the lines. 4*.8 = 3.28, and backwards 5*.8 = 4, so you should be able to talk to both 3.3V and 5V targets without changing, let the resistors and input diodes take care of the rest. Have the ICD and targets right in front of me, I'll try it in just a little and make sure it's good, could be handy and almost no extra parts..

Hi Alan96
Yes a nice turn of phrase
lame crap
... from you removed post.

So I was right - you have NOT tried it on 3v3 devices - re: thread name.
Having to change resistor values for each Vdd is NOT an acceptable design.

As you do not have a PIC24 device - try with 877A externally powered at 3v3 or even go to 2v0 - still within programming spec. ICD2 can do all serial PICs at all voltages.

9 res + 6 trans = 36 holes = high production cost + mega error opportunities.

regards ... Polymath
 

mplab icd2 datasheet mclr

polymath said:
Alan69 said:
polymath said:
Hi szlovak
I do not think Alan69 has tried his own suggestion.



I use it all the time, but of course on normal inputs not schmitt trigger.


I don't use PIC24, but assuming it's inputs are schmitt trigger too, I would try dropping ICD to 4.1V and inserting just series resistors into the lines. 4*.8 = 3.28, and backwards 5*.8 = 4, so you should be able to talk to both 3.3V and 5V targets without changing, let the resistors and input diodes take care of the rest. Have the ICD and targets right in front of me, I'll try it in just a little and make sure it's good, could be handy and almost no extra parts..

Hi Alan96
So I was right - you have NOT tried it on 3v3 devices - re: thread name.
Having to change resistor values for each Vdd is NOT an acceptable design.

As you do not have a PIC24 device - try with 877A externally powered at 3v3 or even go to 2v0 - still within programming spec. ICD2 can do all serial PICs at all voltages.

9 res + 6 trans = 36 holes = high production cost + mega error opportunities.

regards ... Polymath

Pullups maybe needing lowering from an estimated design value for higher speed doesn't suggest anything about then having to change them for every Vdd. Those are only pullups, that circuit would interface 3V and 5V and 3V and 10V without changing resistors.

Again, it ONLY needed testing to make sure ICD2 wasn't expecting 'not high' when nothing was driving. Passes data fine. I tried the passing data through it, and ICD2 does fine. That it can handle 5V-5V or 3V-5V is already known from the article and previous tests. I just did a quick test to test what needed proving, redundant info wasn't on the agenda. Clear from the circuit it'll do about anything, sometimes a rose is a rose.

Brilliantly simple how that circuit goes together plus takes out some of the other parts in the ICD2, and no need for buffer chips for anyone building it from parts. I've got two copies on two different breadboards and it's pretty much a zero chance for error, I take it you didn't build it before deciding it was too much.. Holes are free where I buy PCBs, not that much for 36 even if you pay. Plus no holes for a buffer chip, not like it's that many more.

It's absolutely too good since it will do more too, I took it out. I'll do it up fully and put up a complete design in the next few days.

Alan
 

eicd2 usb

Alan69 said:
Again, it ONLY needed testing to make sure.........
I think NASA may have said this just before the explosion!

Alan69 said:
Clear from the circuit it'll do about anything.........
Curious - anything? sounds like precision engineering!

Alan69 said:
Holes are free where I buy PCBs, not that much for 36 even if you pay. Plus no holes for a buffer chip, not like it's that many more
36 holes x 100k products - na not much .... NOT! A MAJOR cost of any PCB production is Holes - fact.

Alan69 said:
I take it you didn't build it before deciding it was too much..
I have built one using resistors and smt - You may have noticed that smt do NOT need holes.

You are obviously aware that MChip do not warrant chips programmed outside their spec.

regards ... Polymath
 

icd2.5usb icdwarn0020

Hi all,
Anyone here modified lothar stolz's ICD 2 (serial port) to support 3.3V ?

I've seen so many ICD2, and I am starting to get confused, which works and which doesnt ? or buggy?

I've constructed stolz's ICD2 clone, and require 3.3V, can I just add output buffer and the appropriate pins to make it work, or do i have to program the 877A with a different firmware ?
 

icd 3v3

Hi!

First of all, congratulations for your work, I think is an incredible work to clone ICD2. I have some questions (sorry if I don't ask with sense...)

Potyo:

I'm seeing your schematics and I have some doubts:

- You conect the power supply of 74HC4066 to Vpp, but seeing the datasheet of 4066, the max power supply supported is 10V, and Vpp can go up to 13V, then we have a little problem, isn't it?

- X4 is a jumper to select what? Because I see that central pin is target Vpp, isn't it? And the other pins are icd2 Vpp (one directly and the other by 4066), then I don't understand very well what is its function (sorry for my poor understanding...).

- What is X2 and X3 meaning? And isn't possible to reduce this jumpers to only one? If always put one or the other, we can have a 15k resistor in parallel with another 15k resistor with a jumper, then when the jumper is not placed, we have 15k resistor, and when the jumper is placed we have 15k resistor in parallel with 15k resistor, then 7k5 resistor. May be is a fool idea, because I don't know what's the meaning of this resistor (I don't see anything in the datasheet, may be I skip this part), but if it isn't so small resistor we can reduce one jumper. May be we can have no placed jumpers and the idea isn't so good...

- There is any limitation in tolerance on any component (1% or it's ok with 5%-10%)??

Sorry for all questions...

Thanks a lot

Regards

togarha

PS: Sorry for my poor english
 

usb icd clone

I think that should be an economical and effective solution for low voltage pic operation.

21_1160520750.jpg


Level shift from 2.2-5 v from target to schmitt trigger levels (~4V) in programmer is done via an open collector buffer (here a single NL17SZ07 smd, but can be any lov voltage open collector buffer). The buffer used sees an high level from 2-2.2 volt, so it's adequate for low voltage pics, I guess up to 2.5 v.

On the other side, the foltage limiting from 5v programmer to low voltage target is done with 2 diodes on Vpp-tgt and the 330R resistors; cheap and safe.

I must admit I had not yet time to test all the thing, but I'll do it on next days.

Ciao

Max
 

usb 3.3v voltage divider

Tested and it does work with 3.3 v pics.
For non-SMD build, the NL17SZ07 single open drain buffer could be replaced with some equivalent hex buffer in PDIP package that can be run with low voltage and tolerates 5 v on output.... pity that I didn't find any. All 7407 variant are either 5v only or in smd package.
A 5V 7407 could also be used, if powered from 5V and not from target VCC; being TTL compatible, it could detect voltages from about 2.5V as logic '1', allowing programming of 3.3v pics; future lower voltage pics (maybe 2.5v ??) would not work.

Ciao

Max
 

icd2 vpp volt

Hi
Please test it at 2V! The original icd2 works with 2V targets.
 

usb on the pic24

potyo said:
Hi
Please test it at 2V! The original icd2 works with 2V targets.

Well, I've got no 2v pics, but the NL17SZ07 buffer does work down to 1.65V, so it should be no problem.
The only problem is that it comes in SMD package only and I don't know any low voltage open collector buffer in DIP package.

If you use a TTL input compatible 5V buffer (74HCT07 ?), it senses logical '1' only from 2.2 - 2.5 Volt; it can be used only with pics starting from about 2.8-3 V.

I think the only viable solution to have it in through-hole would be to replace the buffer with a 2-transistor discrete equivalent, or, maybe better, to use a single voltage op-amp or even better a comparator with small hysteresis. That would allow sub-1V programming.

What does make my solution cheap and easy is to separate the 2 i/o lines from 16F877A chip and to buffer ONLY the input one. That avoids the use of three state buffers to select data direction.

I did the UP level translation with NL17SZ07, but it can be replaced with some other means. The DOWN level translation is done by resistors and clipping diodes; cheaper than that I thing it's impossible :)

Ciao

Max
 

pic24fj & mclr reset button

Hello!
My first post here:). Anyway, to fix 3v3 problem i used simple level translator made of two NPN transistors (i used some 2N3904) and four 10k resistors - schematic and description is here https://www.edn.com/archives/1996/110796/23DI_02.htm

It works, i used one bidirectional for PGD and a 'half' translator for PGC and MCLR. So it costed me four NPN transistors and eight 10k resistors. Result? It works, tested with pic 24h, and because i use voltage from pic for lower level side i can use it to program 5v and 3v3 pics without changing anything.
I don't think you can get any cheaper than this. And probabily most of us allready have all the components at home.
 

mplab icd 2 24f

Ok, what to do to program 3v3 or lower devices.

- make a level translator from programmer voltage to device voltage. Bidirectional for PGD and PGC (reason for PGC - done some disassembling of 877A code and found BTFSC PORTC,3 lines - looks like in some cases ICD2 tests PGC line to see it's state - my guess is it's checking to see if the line rises fast enough)

- wire programmer so the output stage gets power from target board. So MCLR has levels 0, target Vdd and module Vpp.

- ICD2 doesn't connect 13v to target, if the target doesn't support 13v on the MCLR. So there's no worry that ICD2 would fry the chip.
 

mclr vpp :low

Ok can we summ all this? Can anyone post link to sch of original ICD2, and to clone which can work with all (3v3) PIC devices?

I want to try to build one of these and it would be great if someone do this please.

Best regards!!!
 

www.ada-vajdaság

ADRENALIN2003 said:
Ok can we summ all this? Can anyone post link to sch of original ICD2, and to clone which can work with all (3v3) PIC devices?
You can find it in this topic:
 

icd2 clone level shift

Hello.

Thank you for all informations on this forum.It's so good

I'm not an expert in ICD2 so please be tolerant.

I finish to make a potyo2 ICD2 programmer.
I'm still confuse on the '24H programming way' and mostly what is good or not.

My programmer works very well with all 8bits controler ( that is easy )
It's works well with 24H series.

I don't use the digital pot ( I still don't know why we need it ).
Can you tell me why it's needed ?

I don't use the transistor level because I've got 74HCT125 and 74HCT126 ( error in the diagram ).
Is it the suitable gates ?

I don't use the MCLR disable feature. ( I don't know why that ? )

I use the only part with the blue flag ( red flag is not equiped ).

My Vpp seems HIGH for the 24H. What do you think about that ?

Is someone see somethink missing in the schema ?

Thank you.
 

icd2 scheme

Hi guys
I have an original ICD2, and I found that it used 74LVC1T45 to translate the logic level.
I have test it by oscillograph and found it works perfect! You guys can search the datasheet of 74LVC1T45 for more info.
 

icd2 18f46k20 vdd

HCT125 and HCT126 work the same way, so why bother?

Added after 1 hours 42 minutes:

i've just tried programming PIC16 with potyo2. the voltages were given by the target, and I tried: 2.1V, 2.3V, 2.5V and 3.3V - absolytely no problems at all, but the fact remains that there is a 2.5V on the VDD line when there is no target. anyway I don't think it is a problem. very soon I'm going to test potyo2 with PIC24H [; - maybe next week.

0x41 0x56 0x45!!
 

clone usb device id

Is it possible? Clones most likely doesn't have a voltage regulator outputs.
 

icd2 clone downloading operating system

IamnotJunk said:
Is it possible? Clones most likely doesn't have a voltage regulator outputs.

True, it increases the cost and complexity. The clones are aimed at home users, students and hobbyists. The majority of PIC users they never leave the 8bit chips anyways. If you need high end there are 5V dsPICs, like the dsPIC30Fxxxx.
The 24F and 33F are 3.3V chips but definitely not aimed at the beginner.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top