Post by thread:[PIC]: ICD has 5.7Vpp

Hi, I've looked in the archives for a solution to get my 'free' icd to work. I'm using P Touzet's PCB with Microchip's values of components. When I try to use it with my target circuit I get the 'error performing requested operation'. When I do a self test it says to check MCLR voltage ICD0x4. The ICD options window displays 5.04VDD and 5.70VPP. I'm trying to figure why VPP is so low. I don't have a scope so I'm not sure what it reaches when it self tests. I've measured it after if fails and I read 3.81V on the MCLR/VPP line. The ICD is powered by my target with a 9Vbatt connected to a 7805. I put a 47kohm resistor from 5V to a 1n4148 diode (I don't have a schottky) into the target's MCLR pin. The ICD's MCLR/VPP line is between the target's MCLR pin and the cathode of the diode. RB6 and 7 are connected just to the ICD. I've read that R1 and R21 are sensitive. When I measure the voltage between R1(10k) and R2(4.7k) I get 1.6V. When I measure between R21(100k) and R22(10k) I get 0.36V. Now the 1.6V looks good to me but I know that the 0.36V maybe meaningless as that section of the circuit might not be controlled from the ICD's f876. Should the ICD be charging that circuit to ~13V all the time or just when you want to program the target? The output from the 7812 on the ICD is 3.17V. The LED I'm using drops 2.039V across so I assume that's close enough to 2V. Also I figure the Max232a works because I can see the voltages in the ICD options window. I measured the V+ and V- at ~9V and ~ -8V. The target f870 is controlling an LCD, and when I power up the circuit I can see the cursor blinking. When I probe for voltages I sometimes get the cursor to scroll across the screen leaving blocks. If I then press the reset on MPLAB's toolbar I can get the circuit to reset. I thought that was something, but that's all my ICD can do so far. Any ideas? Anything I should check? Can I self test it without a target circuit? Thanks for reading Brendan -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

[I think I sent the last one as html text, sorry] Hi, I've looked in the archives for a solution to get my 'free' icd to work. I'm using P Touzet's PCB with Microchip's values of components. When I try to use it with my target circuit I get the 'error performing requested operation'. When I do a self test it says to check MCLR voltage ICD0x4. The ICD options window displays 5.04VDD and 5.70VPP. I'm trying to figure why VPP is so low. I don't have a scope so I'm not sure what it reaches when it self tests. I've measured it after if fails and I read 3.81V on the MCLR/VPP line. The ICD is powered by my target with a 9Vbatt connected to a 7805. I put a 47kohm resistor from 5V to a 1n4148 diode (I don't have a schottky) into the target's MCLR pin. The ICD's MCLR/VPP line is between the target's MCLR pin and the cathode of the diode. RB6 and 7 are connected just to the ICD. I've read that R1 and R21 are sensitive. When I measure the voltage between R1(10k) and R2(4.7k) I get 1.6V. When I measure between R21(100k) and R22(10k) I get 0.36V. Now the 1.6V looks good to me but I know that the 0.36V maybe meaningless as that section of the circuit might not be controlled from the ICD's f876. Should the ICD be charging that circuit to ~13V all the time or just when you want to program the target? The output from the 7812 on the ICD is 3.17V. The LED I'm using drops 2.039V across so I assume that's close enough to 2V. Also I figure the Max232a works because I can see the voltages in the ICD options window. I measured the V+ and V- at ~9V and ~ -8V. The target f870 is controlling an LCD, and when I power up the circuit I can see the cursor blinking. When I probe for voltages I sometimes get the cursor to scroll across the screen leaving blocks. If I then press the reset on MPLAB's toolbar I can get the circuit to reset. I thought that was something, but that's all my ICD can do so far. Any ideas? Anything I should check? Can I self test it without a target circuit? Thanks for reading Brendan -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

----- Original Message ----- From: Alan B. Pearce <spam_OUTA.B.PearceTakeThisOuTRL.AC.UK> To: <.....PICLISTKILLspam@spam@MITVMA.MIT.EDU> Sent: Thursday, December 05, 2002 4:22 AM Subject: Re: [PIC]: ICD has 5.7Vpp > >When I do a self test it says to check MCLR voltage ICD0x4. > >The ICD options window displays 5.04VDD and 5.70VPP. I'm > >trying to figure why VPP is so low. I don't have a scope > >so I'm not sure what it reaches when it self tests. I've > >measured it after if fails and I read 3.81V on the > >MCLR/VPP line. > > What sort of inductor did you use ?? I found I had problems using a small > surface mount inductor as it had too high series resistance. I moved to one > designed for use with a switching regulator (Toko D10F series) and now mine > works fine. > > -- > http://www.piclist.com hint: The PICList is archived three different > ways. See http://www.piclist.com/#archives for details. > I used the one from the parts list at http://piclist.com/techref/piclist/freeicd/allparts.htm . JW Miller's 78F331J or digikey part M7843-ND. All my parts are through hole. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

----- Original Message ----- From: samo benedihih <samo.benedicicKILLspamVOLJA.NET> To: <.....PICLISTKILLspam.....MITVMA.MIT.EDU> Sent: Thursday, December 05, 2002 9:39 AM Subject: Re: [PIC]: ICD has 5.7Vpp > Hi! > > I also had some problems getting my ICD to work. I also used P Touzet's > design. There was this problem: faulty transistor Q5. Transistor looked good > when I measured it with multimeter, but there was not enough voltage on > junction of Q5, CR4 and L1 at programming or at selftest (approx.19V, but > that's from top of my head). Problem went away after changing Q5. 13V is > applied to target circuit only when programming is in progress. > Regards, Samo Thanks for the tip. You helped me find one problem with my circuit. I looked at Q5 and was about to pull it when I noticed I had it in backwards. I'm using PN2222As and a PN2907A instead of 2N2222As and a 2N2907A. Are the emitter/collector pinouts reversed between these types? Mr. Touzet's board layout showed the transistors mounted in a certain way. When I assembled it I assumed it was correct. Monkey-see-monkey-do. From what I can tell on both PN2222A and PN2907A transistors have EBC pinout when the face of the transistor is towards you. So I looked at the circuit again and flipped all four transistors. Powered it up again and I _still_ get the same problem! I then replaced all four with new transistors and the same thing. Well one thing did change, the VPP voltage now reads ~3.2V on the ICD options window. It was 5.7V. I'm going to go over the whole circuit again and compare it to Microchip's schematics. Maybe something else is backwards. I just wish I had a scope or a faster multimeter. Anyone else -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

----- Original Message ----- From: samo benedihih <samo.benedicicspam_OUTVOLJA.NET> To: <@spam@PICLISTKILLspamMITVMA.MIT.EDU> Sent: Friday, December 06, 2002 4:03 AM Subject: Re: [PIC]: ICD has 5.7Vpp > That's odd. I just checked datasheets for PN2222A and PN2907A on Fairchilds > site and pinout is ( or should be ) E B C if transistor is faced towards > you. But my Q3 is PN2907A (others are metal can package, so it's easier to > find pinout) and it's inserted like it's drawn in implantation, which is > definitelly wrong according to this part datasheet, and it still works. So, > Q3 must have emitter towards C9, Q5 towards R27 and Q2 and Q4 towards the > edge of the board. Hope I didn't confuse you too much. > Best, Samo > > -- > http://www.piclist.com#nomail Going offline? Don't AutoReply us! > email KILLspamlistservKILLspammitvma.mit.edu with SET PICList DIGEST in the body > It took a few minutes to get my head straight but I now have the transistors in the proper way (according to Microchip's schematics and Fairchild's datasheets). I also found a slight problem with the 0.1uF cap from pin2 on the MAX232A. The Touzet PCB has the capacitor from pin2 to ground. The MAX datasheet shows it from pin2 to VDD. It wouldn't have much to do with VPP being so low but I switched it anyways. After that it still doesn't want to work. I then pulled R21 and substituted a 56k resistor with a 100k pot (used pins 1 and 2 of the pot). I set the equivalent resistance to 100k and tried it again. I then tried the self test for various pot settings and no luck. I guess I'll have to pull all the parts off and test them on a breadboard. So, if I do ever get it to work I should be able to see ~13V using a multimeter on the MCLR/VPP output line right? The 13V signal should stay long enough to read it? Or is a scope the only way? Thanks, Brendan -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! email RemoveMElistservTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body

Caps must be inserted as on P Touzats picture. I have 1micro electrolytic caps. There is an error in Microchip's schematic regarding cap orientation. One question: does MPLAB see your module? If yes, there's probably no problem around MAX232, since it only converts TTL to RS232 logic levels. My R21 is 100k, no problem here.So I measured voltages using very cheap digital multimeter (10$), so you don't need oscilloscope: idle mode( no programming ): voltage on base of Q5 : 0,03V pin2 of 876: 0,03V input to 7812: 4,4V output of 7812: 3,7V junction of Q5, CR4 5V programming: base of Q5: 0,3V pin2: 1,44V input to7812: 15V output of 7812: 10-13,5V junction of Q5, CR4 4,8V I measured these voltages because pin2 modulates signal on base of Q5 in order to produce 21V on input of 7812 and then 13V on output. You should measure voltage on pin2 and on base first. Then you will know at least if your 876 produces sygnal that drives Q5. If you have signal on pin2 and no signal on base, you have open R29 or shorted R28. If you have that signal, check voltages on junction and on input to7812. 5V on junction-check L1 and Q5. Correct voltage on junction and input-check7812. Correct voltage on junction and nothing on input of 7812- open or reversed CR4, maybe shorted C10. Voltage on middle pin of 7812 should be 0,75V. Let me know if that helps. If this part of circuit is OK, then I' ll measure also circuit arround Q2 and Q3. Regards, Samo {Quote hidden}> > So, if I do ever get it to work I should be able to see ~13V using a > multimeter on the MCLR/VPP output line right? The 13V signal should stay > long enough to read it? Or is a scope the only way? > > Thanks, > Brendan > > -- > http://www.piclist.com#nomail Going offline? Don't AutoReply us! > email spamBeGonelistservspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body > > -- http://www.piclist.com hint: To leave the PICList TakeThisOuTpiclist-unsubscribe-requestEraseMEspam_OUTmitvma.mit.edu

----- Original Message ----- From: samo benedihih <RemoveMEsamo.benedicicTakeThisOuTVOLJA.NET> To: <PICLISTEraseME.....MITVMA.MIT.EDU> Sent: Saturday, December 07, 2002 11:44 AM Subject: Re: [PIC]: ICD has 5.7Vpp > Caps must be inserted as on P Touzats picture. I have 1micro electrolytic > caps. There is an error in Microchip's schematic regarding cap orientation. Are you sure? I also looked at the Maxim datasheet and it showed the same orientation. Well, I'm using ceramics so it isn't as big a deal for my circuit. > One question: does MPLAB see your module? If yes, there's probably no > problem around MAX232, since it only converts TTL to RS232 logic levels. My > R21 is 100k, no problem here.So I measured voltages using very cheap digital > multimeter (10$), so you don't need oscilloscope: MPLAB did see my module. It also reported that I didn't have it plugged into a target (which was correct). Thanks to your help I now have it working! I'm getting ~14V on VPP. Thank you. Here's my numbers for the voltage test.... idle mode Q5 base 0.00V pin2 1.6V** 7812 input 4.4V 7812 output 3.12V Q5 junction with CR4 4.99V During Self test Q5 base 0.00V pin2 1.6V** 7812 input 4.4V 7812 output 2.76 - 3.12V Q5 junction with CR4 [forgot to measure it] ** I assumed after looking at your description below that you were in fact talking about pin13/RC2 as it was the pin connected to the base of Q5. I can see how you might have mistaken it for pin2 when your looking at the bottom of the PCB to test the traces ;-). So, I measured pin13 and found this: idle mode 0.00V during self test 0.00V ! There was the problem. Q5 was never switched to create Vpp. So I pulled the micro controller and R28 and R29. The resistors checked out okay and the continuity of the traces were good. So I decided it was the f876. I made a program to turn an LED on and off on RC2. It worked so I then figured it was the code. I reprogrammed the f876 with the mpl876.hex file and made sure to follow the instructions in the ETN-21 note. When I originally programmed the 876 I think I must have missed the ID locations and EEprom check boxes. Now it self tests perfectly. {Quote hidden}> > idle mode( no programming ): > > voltage on base of Q5 : 0,03V > pin2 of 876: 0,03V > input to 7812: 4,4V > output of 7812: 3,7V > junction of Q5, CR4 5V > > programming: > > base of Q5: 0,3V > pin2: 1,44V > input to7812: 15V > output of 7812: 10-13,5V > junction of Q5, CR4 4,8V > > I measured these voltages because pin2 modulates signal on base of Q5 in > order to produce 21V on input of 7812 and then 13V on output. You should > measure voltage on pin2 and on base first. Then you will know at least if > your 876 produces sygnal that drives Q5. If you have signal on pin2 and no > signal on base, you have open R29 or shorted R28. If you have that signal, > check voltages on junction and on input to7812. 5V on junction-check L1 and > Q5. Correct voltage on junction and input-check7812. Correct voltage on > junction and nothing on input of 7812- open or reversed CR4, maybe shorted > C10. Voltage on middle pin of 7812 should be 0,75V. Let me know if that > helps. If this part of circuit is OK, then I' ll measure also circuit > arround Q2 and Q3. > Regards, Samo Thanks for taking the time to help. Now I'm going to try the ICD out on a circuit. Brendan -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestmitvma.mit.edu