PicKit 2 FULL with TH components - Need help with layout

[mvs sarma]

Microchip has a latest version of MPLAB Ver8.2 that works with .net and PICKIT2 ver2.6 as already indicated.

 

[nishal]

@Computerman29 can u post the working of your pickit2.

Whether calibration is successful and working status on 'troubleshooting' menu. This will help us to analyze further.

Nishal

Hi Nishal,

I used yr layout & its fully function. Pls check previous post. However, it can't be calibrate.

Trouble shooting all passed.

Thanks for the nice layout.

Regards

What does it show on calibration? can u post actual DMM voltage and the error during the calibration phase!

Nishal

 

[me_guitarist]

Hi Nishal,

It can't be calibrate, measured voltage is only 0.157V. error msg:'Could not fully calibrate the unit. The USB voltage may be too low to completely calibrate."

I removed Q1 & shorted its D&G(as sugguested by Sarma), give yellow LED (after 470R) a +5V. The kit is now fully working. except the calibration.

 

[nishal]

Thanks Me_guitarist, If it is not calibrating and voltage measures at 0.157v then how come troubleshooting window passes the test!!

Please post your volts at troubleshoot window (It should be approx 4.5v depends on VDD_USB and since Q1 is short, it will be same even if you change the volts at troubleshoot window)

Nishal

 

[wizpic]

Looking at the shcematic Q1 is your mosfet which is fed by the op-amp, in my thinking and understanding it could be be either the mossfet is not supplying the variable voltage eihter the drive current is not enough or it could be driven to hard casuing the mosfet to be fully switched on and not making the voltage variable or you could of chosen the wrong op-amp.

If I'm looking at it correctly the op-amp is supplied with PWM from the pic which this gives you the variable voltage if this is the case then to me it's your op-amp is wrong. Now I could be totaly wrong and these are just my thoughts what could be casuing the problem, Why not try a LM358 or another type of op-amp other than a 0741, To me the TL081&0741 are very old hat and very nosiy op-amps for the modern circuits, Yes in the early days they may have been a good op-amp

My version of the circuit is completed and I have got it 90% built just got to get 2 P channel mosfets which should be here shortly then I can try my version out but I shall try a LM358 first (has my version is drawn for that op-amp or similar foot print for some more op-amps I've got) Then I'll let you know how I get on, I can't see why not the LM358 would not work has I've seen it used on the ICD2 doing the same thing,

 

[nishal]

Wizpic, I tried both LM358 & TL081 and shows same result. Currently using TL081.

 

[me_guitarist]

Thanks Me_guitarist, If it is not calibrating and voltage measures at 0.157v then how come troubleshooting window passes the test!!

Please post your volts at troubleshoot window (It should be approx 4.5v depends on VDD_USB and since Q1 is short, it will be same even if you change the volts at troubleshoot window)

Nishal

Sorry Nishal,my mistake.

-0.157V is when fitted with Q1 & D&S doesn't shorted.

when Q1 removed, D&S shorted result at trouble shooting:
Verify VDD=5.0V
Verify VPP=11.9V

I can't verify PGC & PGD as I don't have oscilloscope.

Regards

 

[nishal]

Dont feel sorry, everybody make mistakes!!
Anyway our target is to solve the problem!

Good volts. You can verify PGC & PGD using DMM. It will show some volts at relevant pins when signals are ON. But for the time being we are not concerned on these signals.

Nishal

Added after 3 hours 3 minutes:

Wizpic, I used 741 not LM358. Sorry for the mistake

Nishal

 

[Computerman29]

Greetings,

It appears the op-amp is a problem for some builders, and I would like to offer my advice.

I will confirm that the unit will calibrate properly and passes all tests using the op-amp in my original schematic. I get about 4.65v with both my meter and in the software - which equals 5v minus about 1 diode drop.

The only substitutes that I have found suitable are from Microchip. The dual version of the original PicKit2 (MCP6002), MCP6273, MCP6281, MCP6283, and the dual versions of those. Also the dual version of the one I used (MCP6272). They all are available in DIP8, and all are about the same price (at least in the USA).

I have done extensive research before I chose the op-amp. I searched Microchip, TI, ADI, Fairchild, National, Intersil, ST Micro, ON Semiconductor, and Linear Technologies. Although others may work, the op-amps stated above are the only suitable replacements I could find that I would recommend. Output voltage swing, slew rate, Vos, CMRR, PSRR, Ib (the Microchip op-amps have Ib=1pA!!!), and noise seem to be very important. It also MUST be single supply, CMOS, and R-R I/O - quite demanding!

Also, you guys are AMAZING!! You have already done much better than myself before your units are even functional!! Your finished product will probably be better than the original!! I sincerely hope you share your final results. It would be much appreciated!!

Good luck to you all,
Computerman29

 

[me_guitarist]

Greetings,

It appears the op-amp is a problem for some builders, and I would like to offer my advice.

I will confirm that the unit will calibrate properly and passes all tests using the op-amp in my original schematic. I get about 4.65v with both my meter and in the software - which equals 5v minus about 1 diode drop.

The only substitutes that I have found suitable are from Microchip. The dual version of the original PicKit2 (MCP6002), MCP6273, MCP6281, MCP6283, and the dual versions of those. Also the dual version of the one I used (MCP6272). They all are available in DIP8, and all are about the same price (at least in the USA).

I have done extensive research before I chose the op-amp. I searched Microchip, TI, ADI, Fairchild, National, Intersil, ST Micro, ON Semiconductor, and Linear Technologies. Although others may work, the op-amps stated above are the only suitable replacements I could find that I would recommend. Output voltage swing, slew rate, Vos, CMRR, PSRR, Ib (the Microchip op-amps have Ib=1pA!!!), and noise seem to be very important. It also MUST be single supply, CMOS, and R-R I/O - quite demanding!

Also, you guys are AMAZING!! You have already done much better than myself before your units are even functional!! Your finished product will probably be better than the original!! I sincerely hope you share your final results. It would be much appreciated!!

Good luck to you all,
Computerman29

I use MCP6271, BS170 & BS250(MOSFET). However, the calibration failed as well.
So what should we do now?

 

[mvs sarma]

Greetings,

It appears the op-amp is a problem for some builders, and I would like to offer my advice.

.....................
Good luck to you all,
Computerman29

I use MCP6271, BS170 & BS250(MOSFET). However, the calibration failed as well.
So what should we do now?

i request you yo indicate the make of BS170/250 FETS. In know that BS250 follows DGS order. In Nishal's PCB, it needs to be swap mounted. If mounted wrongly, the internal diode takes over with loss of control. we get the out put at .6V less than the Vusb and no control by op amp. things might work but no calibration.

 

[mvs sarma]

@Me_guitarist,
try to get a costly part like NTD20p06L or a cheaper but smd part IRLML6402 and replace at Q1 location with care of proper pinning. anyway that you have used MCP6271, you job will be done.
Farnel is able to provide them

Compurterman29 has once indicated that BS250 can't replace NTD20P06 neither it could work for IRLML6402.

Sarma

 

[me_guitarist]

@Me_guitarist,
try to get a costly part like NTD20p06L or a cheaper but smd part IRLML6402 and replace at Q1 location with care of proper pinning. anyway that you have used MCP6271, you job will be done.
Farnel is able to provide them

Compurterman29 has once indicated that BS250 can't replace NTD20P06 neither it could work for IRLML6402.

Sarma

Thanks Sarma,

I'll try to get the part & post up the result.
BTW, do you tried?

 

[mvs sarma]

@Me_guitarist,
try to get a costly part like NTD20p06L or a cheaper but smd part IRLML6402 and replace at Q1 location with care of proper pinning. anyway that you have used MCP6271, you job will be done.
Farnel is able to provide them

Compurterman29 has once indicated that BS250 can't replace NTD20P06 neither it could work for IRLML6402.

Sarma

Thanks Sarma,

I'll try to get the part & post up the result.
BTW, do you tried?

Ordering for IRLML6402

 

[me_guitarist]

Ordering for IRLML6402

Hi Sarma,
Do you success using IRLML6402?

 

[Computerman29]

Greetings,

Compurterman29 has once indicated that BS250 can't replace NTD20P06 neither it could work for IRLML6402

I do not recall saying that, but it seems to be confirmed with people's experiments. I have not used either so I don't know. They could possibly be used if you make sure the pinout is correct!! I cannot stress that enough!! The only way to be absolutely sure the pinout is correct is to manually force route it. In the schematic, disconnect all pins on one MOSFET, then connect ONE pin and check the layout to see if it matches. Repeat for the other pins and the other MOSFET's. That is how I did mine. None of the symbols were correctly routed to the footprints. That is why I had problems at first, and also why I was able to get mine working with an obviously incorrect schematic and layout. I am not trying to imply that anybody is stupid (quite the contrary!), but I am just making sure that I stated that properly.

 

[yus]

I just want to add some info.

I was able to build a PICkit2 clone with these particular devices:
transistors:
NPN -> BC817
PNP -> BC807
NMOS -> IRLML2803
PMOS -> IRLML5103
diodes: BAT54 and 4148
opamp: TLV2372

all of them are surface mount devices.
finished circuit: **broken link removed**
*it has no eeproms

 

[Jose Fausto]

Hi friends
Please check our PCB for PICKIT2, only uses SMD parts for FDC6420, ZHCS1000,IRLML6402, BAT54, and MCP6001, all other is DIP, of course, the PCB, was made for comercial use, and made for electronicahifi in Peru

 

[mvs sarma]

Greetings,

......
I do not recall saying that, but it seems to be confirmed with people's experiments. I have not used either so I don't know. ........

Agreed that you did not say that. It was said by me_guitarist. Sorry for that CM29!

 

[me_guitarist]

any news on this project?

My board work very well since I built.