[SOLVED] Project to replace CY7C64613 in the ICD2

[potyo]

Second idea: short connection or break on the wires between 877 and 4550.

 

[Computerman29]

A question about Potyo's design:

Is it ok to use 7407 instead of 74LS07? Other than the power consumption, what is the difference between the two?

 

[Gandalf_Sr]

Still struggling to get mine working. I'm currently running USBmon but I'm struggling to understand it. Can someone take a look at this screenshot...

I think that tells me that the 4550 is talking to my PC, can someone confirm that please?

I did a data log but didn't see any packets from 0x82 but maybe I was doing something wrong?

Is there a way that you can send a specific message to the 4550 and look at the reply it gives?

I've been through the PCB under a BIG magnifying glass and tested for shorts. My thoughts are that maybe I can diagnose the issue from reading the USB data... with a little help from my friends

Added after 3 hours 6 minutes:

I've been doing some thorough checking around my non-working board (Potyo's Vn2) with many ICs removed.

I've had no more input yet but I think that the USB is communicating (see earlier post with screen capture from USBmon). Indications seem to suggest a hardware or software issue with the 877A.

Software Thoughts
Someone called Jim sent me a message saying

For me the problem was in the 877a in that it could not comunicate with the 4550 properly using the boot program, not sure of the exact file off the top of my head. Also not knowing any more about the boot issue I tried the 661.hex and it worked.
Hope this helps.

but I don't understand which file he's referring to when he says that he used the "661.hex file" to program his 877A.

Harware Thoughts
One thing that concerns me is to do with the 3 way jumper. Potyo says to set this so that it's nearest IC5, the 74HC4066 but that means that the full VPP (up to 14V) is sent into pins 5, 13, 12, and 6 of the 74HC4066 via CMOS-EN. I have used a 74HC4066E and I looked at the data sheet. The Max DC supply voltage for this device is 10.5 V so it looks like this IC may being sent too many Volts. Should I be concerned? Should I put a resistor from the 3-pin jumper to the 4066 input pins? Have I used the right device? Maybe there's a different version that handles the extra Volts?

I checked that the VPP inverter was running. I removed the digital pot chip and put a 10k pot between pins 5 and 6; watching VPP while I turned the pot, it ranged between 10.4 and 14.2 Volts - I think this is good. FYI I used a 150uH inductor (see picture of my board in earlier post).

I checked the operation of MCLR for the 877A via Q4 and it works fine - ranging between 0.6 and 4.89 V when you take Pin 36 of the 4550 to +5v or ground.

Also, there is no connection to Pin1 of J3 but, depending on the setup for a target board via the ICSP I believe that pin 6 might be needed. Is this correct?

Has someone reading this actually made the Potyo Vn2 board as shown on Page 28 of this thread and made it work? I downloaded the Eagle files and printed the PCB directly from Eagle CAD. I also ran the Electrical Rules Check and it seems to say that all is OK, at least the schematic and the board are consistent.

Once I have mine working, I will publish a full set of documentation on how to trouble shoot this puppy.[/quote]

Added after 1 hours 43 minutes:

OK, I now realize that this will likely get glued on the end of my previous post but I wanted to tell y'all that.....

I MADE IT WORK!!!!!
The problem was Q4 but there were several things wrong...

I must be going crazy in my old age, Q4 is a BC237, Q1-3 are BC327s but I got it into my tiny brain that they were all BC327s. BIG mistake as BC237 = NPN and BC327 = PNP.

As I said above

I checked the operation of MCLR for the 877A via Q4 and it works fine - ranging between 0.6 and 4.89 V when you take Pin 36 of the 4550 to +5v or ground.

Well that was true but only when there were no ICs in the sockets and I have no idea why it worked in that situation. By chance I tested Pin 1 of the 877A with a meter and found it was at 6.3 mV, clearly it needed to be at +5V. I removed the 4550 but left the 877A in place and tried to make Pin 1 MCLR go between 5 and 0 v by connecting Pin 36 of the 4550 IC socket to 0 and 5 v but Pin 1 of the 877A stayed low which started me looking very closely at Q4 and the surrounding components.

Anyway, I put a BC237 in as Q4 but in triple-checking everything I realized (as has been said earlier in this thread) that Q4 is shown in the wrong orientation in Potyo's V2 Eagle CAD board file; the flat side should face the 4550, not the 877A.

so... a BIG THANKS to Potyo and all the others that made this possible. Here's a summary of my Output data....


Auto-connect not enabled - Not connecting (Try enabling auto-connect on the ICD2 settings pages.)
Connecting to MPLAB ICD 2
...Connected
ICD0082: Failed MPLAB ICD 2 operation
MPLAB ICD 2 Ready
Running ICD Self Test
ICD0082: Failed MPLAB ICD 2 operation
Running ICD Self Test
ICD0082: Failed MPLAB ICD 2 operation
Connecting to MPLAB ICD 2
...Connected
ICDWarn0054: MPLAB IDE has lost communications with the MPLAB ICD 2. Would you like to attempt to reconnect?
Connecting to MPLAB ICD 2
...Connected
ICD0286: Unable to communicate with ICD
ICD0082: Failed MPLAB ICD 2 operation
MPLAB ICD 2 Ready
Downloading Operating System
Connecting to MPLAB ICD 2
...Connected
Setting Vdd source to target
ICDWarn0020: Invalid target device id (expected=0x4D, read=0x0)
...Reading ICD Product ID
Running ICD Self Test
... Failed Self Test. See ICD2 Settings (status tab) for details.
...Download Operating System Succeeded
MPLAB ICD 2 Ready
Running ICD Self Test
... Failed Self Test. See ICD2 Settings (status tab) for details.

It says in the status window that the target MCLR Vdd is low but I think that's cos there's no target connected and I've set it so that the target supplies its own Vdd???

My question about the HC4066 voltage and the 3-pin jumper above still stands; Potyo???

 

[jim2098]

I'll try to fill in some of the blanks to my orginal message and elaborate on the things I noticed upon getting up and running.
If you have loaded the "Microchip" IDE drivers as per the tutorial in the "Microchip" lab properly, and you plug in the ICD2, while in the device manager you should see "Microchip" logo popup. This happens every time and it means that there is some communication from pc to 4550. This will allow you to separate your areas of where the problem lies.
If you don't get the logo popup in device manager then it's a hardware problem or
the "IDE' is not loaded properly.
If you get the logo then I would look at the 877 or 877a. The jumper won't make any difference until you connect to program or debug.
The bootloader I used is called "ICD661.hex", it should be on line.
Hope this helps.

 

[potyo]

My question about the HC4066 voltage and the 3-pin jumper above still stands; Potyo???

There are several types of 4066 switches. You can use for example: M74HC4066B1 from ST Microelectronics, MC74HC4066 or MC14066B from Motorola or the old CD4066 from any manufacturer.

 

[Gandalf_Sr]

Jim2098.

Thanks for your input. If you read my post above you'll see that I found the problem which was simply that I screwed up and put in the wrong Q4 transistor. The effect of that mistake was that the 4550 was not able to execute a MCLR on the 877A and my symptoms were, as several people suggested, effectively due to a problem of communication between the 4550 and the 877A. For anyone thinking of building an ICD2 clone, I can state that Potyo's Vn2 board works with the files provided. The only error that I know of right now is that Q4 is shown in the wrong orientation on the layout. It's not an error but the RJ12 connector that the board is designed for is not readily available. Once I'm sure that there are no other issues, I'll fix the Eagle files and post them up here.

Added after 6 minutes:

Potyo, My worry about your 3-link jumper system is the fact that in one position it feeds VPP directly to 4 inputs on the 4066 and, at 13V for VPP, that this exceeds the max input volts for the 74HC4066. If I'm right it could be easily fixed by a diode or resistor between the center link of the jumper and the 4066 inputs.

 

[Gandalf_Sr]

My question about the HC4066 voltage and the 3-pin jumper above still stands; Potyo???

There are several types of 4066 switches. You can use for example: M74HC4066B1 from ST Microelectronics, MC74HC4066 or MC14066B from Motorola or the old CD4066 from any manufacturer.

Thanks Potyo.

I think the point to bring out here for would-be builders of Potyo's V2 board and the answer to my earlier question is that not all versions of the 74HC4066 can handle the full VPP voltage (so RTFM). Choose your version with care; the original CD4066 is good for up to max 15V supply, I'm gonna order a different 4066 than the 74HC4066E I currently have which is only good for 10.5V.

Potyo, I realize that you don't like repeating information but I have now read all of this thread. For the benefit of me and other would-be builders of your V2 board, PLEASE can you explain the intended use of the X5 3-pin jumper on your V2 board?

 

[Beowolf]

It is a two-state jumper.
In one position, it is an ordinary ICD2, which means that it is permanently connected to PGD and PGC on the target chip (this is when VPP is permanently connected to 4066). In other position it is connected to the target only when Vpp is applied to MCLR pin of the target chip, allowing you to use these two pins in your project (wery helpful on 12F pics). Unfortunately, debugging is done without Vpp, so in this position you can't debug the target, only program.

 

[ender84567]

I'm fairly new to eagle cad, I've started with pyoto's v2 file and am making several changes, including going to QFP for the pic's, is there anything other than the orientation of q4 that is wrong, i have read the whole fourm, but somehow i missed that orientation issue untill this last page, and i may have missed others. I'm replacing a few other ic's as well with things i have laying around the house (in true homebrewers spirit). I'll post gerbers and probably the eagle file when i'm done (if the one i build works), anyother good ideas from the peanut gallery on this? i'll read over the forum again and try and be more carefull in my documentation of known bugs, hopefully i can machine a board this week and let you guys know how it goes, thanks for any future and all help up to this point.

 

[folks]

is there anything other than the orientation of q4 that is wrong

Yes, it is wrong, see the Q4 datasheet and compare with the potoyo's layout.

Somebody could say me if the MC34063 works well along with MCP41010, because I have found some differences with respect to the original one.

Varies the VPP from 3.3v to 13v?

Thanks!!!

 

[togarha]

Hi!!

The L varies (with the C1) the oscillator frequency, then It doesn't matter so much with it.

About the resistors, it doesn't matter too, because the dpot. Dpot is who set the VPP output, then, it can be working in other ranges, but it will work fine...

regards,

 

[Gandalf_Sr]

I tested mine. I put in a 10k pot between on the DPOT connection in your diagram and turned it end to end. My Vpp ranged from 10.2 to 14.8 V and my L is 150uH

 

[potyo]

The oscillator frequency is fixed, not depends on L or C1.

 

[togarha]

oscillator frequency (of mc34063) is not fixed, you can see Ct & Lmin formulas at datasheet

Regards,

Added after 2 minutes:

Gandalf in which of both schemes do you put the pot?

 

[Pierrre]

Hi all,

First of all thanx for your works !

I've made the ICD2 Clone Rev, i'd to made an adaptor cause the connector RJ12 is little different in France. Off course i checked all connection and all componants values and they match Clone's schematic.

I have ICD2 connection but no target recognition... I tried on a 16f877.

I measure 3.12v beetwen pin 3(gnd) and 5(PGC)
I measure -1.89v beetwen pin 3(gnd) and 4(PGD) on the RJ 12 connector on the clone. Ov on all others ...
Is it the good measure ?


I have this message in MPLAB :

Connecting to MPLAB ICD 2
...Connected
Setting Vdd source to target
ICDWarn0020: Invalid target device id (expected=0x4D, read=0x0)
...Reading ICD Product ID
Running ICD Self Test
... Failed Self Test. See ICD2 Settings (status tab) for details.
MPLAB ICD 2 Ready

Maybe my adaptator wrong ...

I need your lights please ! :idea:

Thanx in advance

 

[crazyduck]

Hi,
Voltages aren't be able to be measure on PGC and PGD because they aren't continuous.
MPLAB measures VPP and VDD voltages. You should be to find between 11V and 13V for VPP and 5V for VDD on the target PIC !!

Salut,
Les tensions ne se mesurent pas sur les broches PGC et PGD car ce sont des signaux variables.
MPLAB mesure les tensions de programmation VPP et d'alimentation de la cible VDD.
Soit entre 11 et 13V pour VPP et 5V pour VDD.

Added after 9 minutes:

Hi,
I've made a new Double sided PCB version from the "pinguoindunord" schematic.
I've test it, and it is functionnal at 100%.

I've included in the archive the "universal programmable module" to connect to ICD2. With this module, you've got an universal programmer for all the PIC in DIP module. The module is based on the Firefly board.

CD

 

[Pierrre]

Hi !

Thank you for your response.
As i understood PGC and PGC varies but i mentionned that as an element for being sure that the problem come from my adaptor ...
So i have the good VPP but vdd of the target is not the good one ... cause i measure 0V on pin 2(VDD) of the connector instead of 5V


Any ideas ?

Thanx in advance

Salut et merci pour ta réponse CrazyDuck !

J'avais compris que PGC et PGD varie afin de programmer la cible. J'avais mentionner mes mesures comme un élément afin de déterminer si le problème venait bien des connections de mon adaptateur.

Ceci dit je ne retrouve sur aucune broche une tension de 5V sur mon connecteur RJ12 (sur le clone)... donc meme si inversion de ma part les tensions mesurée sont mauvaise...

As tu une idée ?

Merci d'avance

 

[folks]

Hi, Thanks to all.
I am a little confused in this part (Dpot).
Dpot automatically adjusts according to the value that has RA3?

Regards!!!!!!

 

[togarha]

folk,

Yes, the pic automatically adjusts DPOT according to the value at RA3...

 

[ender84567]

@potyo, how do you get such a nice looking board out of eagle? i used your schematic, made some changes, added an extra pin header, and i used the autoroute (using almost an identical placing) and i got something very ugly looking, how is yours so awesome?