C7005 chip repair - well almost - help please?

3786 Views 75 Replies 6 Participants Last post by Dr_C, Nov 28, 2020

slotterkeith

Discussion Starter · Oct 19, 2020 (Edited by Moderator)

Hello.

re. C7005, PIC16F630

INITIAL ERROR
While PIC pin 10 output is 1.7v there is no continuity to bases of TR3 and TR4 as circuit diagram.

(CAUSE
This follows intermittent operation and subsequent chip failure. Trace assumed broken when altering chip position and messing with car bodyshell.)

ACTION
Soldered 30AWG jumper from pin 10 to TR3 / TR4 bases. No shorts.

NEW ERROR
Now there is no output from pin 10.

OBSERVATION
I noted earlier that pin 9 gave exact same output voltage as pin 10. It still does (since jumper fitted).
I have also found that pin 9 and 10 spec is identical on datasheet.

QUESTIONS

1: Can I use pin 9 as an alternative? Or will it not have been programmed?

2: What might I have done to cause output volts failure to pin 10?

3: If pin 9 cannot be used in lieu of pin 10, could the chip be repaired by re-flashing with In Car Pro?

The last question is academic at least until I learn Assembly Language or whatever it takes.

Thank you. I intend to learn from any mistakes - however long I spend.

See less See more

1 - 20 of 76 Posts

Dr_C

· #2 · Oct 19, 2020

Appreciate I am not answering your question but... I suggest the following...

1/ check the regulator is outputting 5V when powered up.
2/ if so then check whether the PIC is running by either:
- using an oscilloscope to see if there is a PWM waveform on pin 7 which drives the IR LED or...
- hand roll a car (fitted with the decoder) past a XLC sensor on a SSD powered-up track and see if the XLC triggers.

If 1 and 2 are successful then fault tracing can move to TR1-TR4 circuits.

c

slotterkeith

Discussion Starter · #3 · Oct 19, 2020

Dr_C said:
Appreciate I am not answering your question but... I suggest the following...
1/ check the regulator is outputting 5V when powered up.
2/ if so then check whether the PIC is running by either:
- using an oscilloscope to see if there is a PWM waveform on pin 7 which drives the IR LED or...
- hand roll a car (fitted with the decoder) past a XLC sensor on a SSD powered-up track and see if the XLC triggers.
If 1 and 2 are successful then fault tracing can move to TR1-TR4 circuits.
c

Thanks ever so much for all your help. (I put this on a separate thread so the the 'jig' one would not get sidetracked).

I have already done both of these tests:

The chip is powering up. The XLC does not trigger under those (push along track running) circumstances. I have not scoped this but will do so to double check.

So as Greg has pointed out in an earlier post, I guess the chip may well be at fault... which leads me into, as yet, 'uncharted territory'. Hence I have been reading up your PICkit 4 thread and watching hours of YouTube. Totally fascinating and enough to keep me busy during lockdown, and well beyond!

I will continue with the physical side of the 'jig' arrangement in due course.

Take care.

K

See less See more

slotterkeith

Discussion Starter · #4 · Oct 24, 2020

I had to have a few days off this project as it was beginning to get the better of me.

Just to say that I have repeated test 2 with the Oscope and there is no voltage and waveform shown.

to be continued...

Dr_C

· #5 · Oct 24, 2020 (Edited by Moderator)

Hi Slotterkeith,

If the 16F630 microcontroller is receiving 5V supply but not outputting an IR LED drive waveform it is very likely the 16F is damaged. There are other possibilities, such as a failed ceramic resonator, but all of these possibilities are a bit unlikely.

If you want to take this further there are two main options:

1/ use a Hornby programmed 16F630 from a donor SSD decoder.

2/ fit a brand new 16F630 and programme with ICP v4.0.1. For reference the Ts and Cs for ICP v4.0.1 state...

<The firmware provided here is done so free of charge and may only be
implemented on car decoders (in-car modules) manufactured by Scalextric/Hornby.>

Clearly loading onto non-Hornby-manufactured decoders would breach these terms. However, fitting a new PIC into a Hornby manufactured decoder where this is to replace a failed one would meet these terms as the decoder is itself Scalextric/Hornby manufactured.

Are you up for replacing the 16F630 and then using a PICKit to programme it?

c

See less See more

GregK

· #6 · Oct 25, 2020

Font Rectangle Parallel Number Paper

See less See more

Reactions: 1

Dr_C

· #7 · Oct 25, 2020 (Edited by Moderator)

GregK said:
ICP Licence.JPG

Rectangle Font Parallel Screenshot Number

Thanks GregK, your text is interesting and clearly is different from the text circulated with In-Car Pro v3.3. For me the interesting change is the added text around derivative work not being called In-Car Pro.

Also, since the licence terms and conditions state that...

<Viewing, downloading and/or use of this firmware implies an unconditional acceptance of these conditions of use and any subsequent amendments made from time to time>

... this raises an interesting question around legitimacy of changing these terms and conditions and whether there is any obligation by the parties making these changes to disclose their changes to those users who received the original terms and conditions supplied with v3.3.

Any thoughts?

c

See less See more

slotterkeith

Discussion Starter · #8 · Oct 25, 2020 (Edited by Moderator)

Many thanks Greg and Dr C.

An interesting observation Dr C. When I was working I would have been obsessed with such copyright intricacy! Thankfully though, here I would simply like to learn how to replace and or reprogram a bad microcontroller by flashing the original, or programming a third-party component with InCar-PRO. This is for two reasons: academic interest / learning a new skill, and to keep my toys running with home repairs as required. NB I would have to learn the procedure more or less from scratch as my only similar experience is having set up an Arduino a few times.

Re. Earlier:

I have just taken a screen shot of the output from pin 7 as mentioned above. It is not entirely dead as you will see. I do get a mean of 44mV. I presume this suggests that the microcontroller does indeed have a problem. If so, yes, my intention would be to replace the 16F630 and then use PICkit to program it. Warning: this is not a pretty sight.. unless of course I have the scope set up incorrectly. I have only bought it a few months ago!

See less See more

slotterkeith

Discussion Starter · #9 · Oct 25, 2020

The idea of users agreeing to unstated issues is interesting. It's a shame my smileys don't work.

"Viewing, downloading and/use of this firmware implies an unconditional acceptance of these conditions of use and any subsequent amendments made from time to time."

Dr_C

· #10 · Oct 25, 2020

Hi Slotterkeith, yes this looks very much like a dead microcontroller... if you would like a suggested approach to removal of the microcontroller without risking damage to the underlying pad areas please let me know - happy to share.

c

Dr_C

· #11 · Oct 25, 2020

By the way... a very nice oscilloscope... dual channel and lots of bandwidth... excellent choice :thumbsup:

c

slotterkeith

Discussion Starter · #12 · Oct 25, 2020 (Edited by Moderator)

Dr C.

The crystal X1:

Before removing the microcontroller I am tempted to take off the crystal so that I can test its tracks for continuity lest there be a circuit trace break similar to the afore-mentioned. As Greg has said X1 has three legs. I have managed to probe under one side and confirm continuity between this centre pin and ground. I may also be able to test the crystal off board, but I cannot read the engraved part no. beyond what looks like E6... Do you know this please so I can attempt a measurement and identify/verify the frequency?

Removing the crystal and microcontroller:

I had envisaged using my cheap, but OK, rework station and air-heating the components to a temperature of 350 to 400 degrees, then gently lifting or flipping them off with tweezers. I would take precautions not to overheat anything, let the heat stray too much, or blow the loosened components away! I might also daub some flux and a spot of fresh solder onto the mc pins beforehand. Would you see this as a suitable approach?

Please notice:
Once again I have a fair knowledge of how this might be done, but my practical experience in these matters in very limited. With respect to the latter, I am developing improved handling techniques, and confidence daily. This in itself is very satisfying. Gratitude.

Keith

See less See more

Dr_C

· #13 · Oct 25, 2020

Sorry a bit busy this evening... the part you mention is a ceramic resonator not a crystal - its 20MHz for the SSD 16F630 devices. A similar part is RS 283-955. I am not sure if this has the same physical dimensions but could certainly be used for testing purposes. That said the ceramic resonators are quite robust and unlikely to fail.

c

slotterkeith

Discussion Starter · #14 · Oct 26, 2020

I will research ceramic resonators! Thanks.

Reactions: 1

Dr_C

· #15 · Oct 26, 2020

Looking again at the earlier scope trace I see the periodicity of the waveform is 20ms i.e. 50Hz so likely its inductive mains pick-up.

c

RikoRocket

· #16 · Oct 26, 2020 (Edited by Moderator)

Dr_C said:
40142E1B-416C-4164-BCED-BA7930A3636B.jpeg

Thanks GregK, your text is interesting and clearly is different from the text circulated with In-Car Pro v3.3. For me the interesting change is the added text around derivative work not being called In-Car Pro.

Also, since the licence terms and conditions state that...

<Viewing, downloading and/or use of this firmware implies an unconditional acceptance of these conditions of use and any subsequent amendments made from time to time>

... this raises an interesting question around legitimacy of changing these terms and conditions and whether there is any obligation by the parties making these changes to disclose their changes to those users who received the original terms and conditions supplied with v3.3.

Any thoughts?

c

I spoke directly to Ian (MIH) about changing InCarPro to make it compatible with the new Rev H chip, he gave permission to use the source code to evolve it and make it compatible, so as V4 was evolved by DRC with Ian's permission I don't think there is an issue with using that name. I suggested the name to Ian in the first place! The only caveat was to replicate the licence terms.

Dr_C

· #17 · Oct 26, 2020 (Edited by Moderator)

Thanks Riko thats very helpful and much appreciated.

On the point concerning licence terms, I can confirm that everyone who has asked for (and duly received) ICP v4.0.1 files from me has also received the licence terms and conditions. These use the text from MIHs original licence with updates where required to ensure ongoing validity. All key principles of the original licence, for example, non commercial use only AND only for use with Scalextric/Hornby manufactured in-car decoders are clearly stated.

Its very important that all users of InCarPro firmware respect these conditions of licence.

c

Ade

· #18 · Oct 26, 2020

2

So on that basis all your new chips have processors for them lifted from existing Scalextric chips

As you say the icp license states firmware only to be flashed onto existing Scalextric chips and I've not seen another new thread stating that firmware is already completed for new type of processors.

Apologies for the drift

See less See more

Dr_C

· #19 · Oct 26, 2020 (Edited by Moderator)

Hi Ade - thanks for the question... and yes you are spot on in noting that ICP cant be used on non Scalextric/Hornby manufactured decoders.

But I am really keen we all keep this thread on the subject of repair of existing SSD decoders - including those where microcontroller failure is suspected. So I will be very brief in addressing Ades post...

1/ no one should be surprised to learn that the intensive decoder hardware development effort currently ongoing is being supported by a similar level of firmware development effort.

2/ Happy to follow up in more detail but suggest that takes place on the link below where I was commenting earlier today on this very subject. The new firmware is called Decoder-Pro and is being developed specifically for decoders based on the 16F1615 and nRF52 series. The new firmware is not being developed for the 16F630 and 16F1503 as used to date in Scalextric/Hornby decoders... so its new decoder hardware with matching new firmware. If commercialisation is to be attempted then both new hardware and new firmware would be prerequisites.

https://www.slotforum.com/forums/index.php?showtopic=197475&p=2392449

3/ now back to decoder repairs...

c

See less See more

Dr_C

· #20 · Oct 27, 2020 (Edited by Moderator)

Hopefully the digression around ICP and licence terms has drawn to positive consensus.... so lets return to decoder repairs... an area where we all, I believe, continue to learn by sharing...

slotterkeith said:
I had envisaged using my cheap, but OK, rework station and air-heating the components to a temperature of 350 to 400 degrees, then gently lifting or flipping them off with tweezers. I would take precautions not to overheat anything, let the heat stray too much, or blow the loosened components away! I might also daub some flux and a spot of fresh solder onto the mc pins beforehand. Would you see this as a suitable approach?

Your approach sounds great... a desoldering rework station sounds good but I guess you will need to take care not to melt the wiring and possibly the plastic of the LED dome. A temperature between 200-250 degrees C should be good - the lower end if 60:40 (Sn

b) solder was used and the higher end if lead-free (more likely).

I tend to use a simple approach using an 18W soldering iron and flooding both sides of the TSSOP package with new solder. Then I alternately heat each side for about 3-5 seconds repeating until both sides are simultaneously molten. Then the chip can be flipped away.

c

1 - 20 of 76 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.