Identifying IC chips that have been sanded down?

All frequent questions about ICs of all kinds, will be stored here.

Identifying IC chips that have been sanded down?

Postby jimosity » 06 May 2009, 13:01

I have a couple units that I want to RE and a couple IC chips have been sanded down to avoid this sort of thing.
What options are there for figuring this sort of thing out?
jimosity
Breadboard Brother
 
Posts: 129
Joined: 12 Sep 2008, 04:23
Location: Cincinnati, OH
Has thanked: 2 times
Have thanks: 5 times

Re: Identifying IC chips that have been sanded down?

Postby Dirk_Hendrik » 06 May 2009, 15:23

In many cases it's heavy to determine the exact type but it will be possible to determine the sort of IC.

Important.
- Think simple. Chances that really difficult, hard to find, alternative components were used is very odd. Probably the manufacturer wants an easy supply chain too. Hard to find components don't help in that case.
- Don't be afraid to make an assumption. Just verify it later. From the tracing after making an assumption you'll get to know tour circuit better so no time is wasted.
- Do you have other schematics from this pedal builder/manufacturer? Use em for reference. Chances are good that blocks of circuit have been re-used or at least the same design topology was applied (Ever noticed that about all bias pullup to half VCC resistors in a Maxon/Ibanez are 510k? When tracing an Ibanez I almost automatically connect them to half Vcc (to verify that later) of that fit's the function.

- For example, in the case you suspect a 8 pin IC to be an opamp of some sort one starts off with identifying it's power supply pins. In the case they are on 4 (ground) and 8 (vcc) chances are good you're dealing with a dual opamp with "standard" pinning like for instance a 4558 or a TL072. Reverese the surruounding components like they are conneted to that dual opamp and see if the circuit makes sense that way (e.g. amplifier stage). If it's pin 4 and 7 that are power supply pins assume it to be a single opamp and see if it matches when a circuit is applied.
- In case of an 14 pin IC one could check if the supply pins are 4 and 11. In that case chances are good you're dealing with a quad opamp.
- Or the other way around, where you're absolutely sure of the function but cannot match that to standard pinning. Is there a component that will match? A common component like a LM339 is a good example (in which the power pins are the other hint btw)
- If dealing with a 14 or 16 pin IC with power supply connected to pin 7 or 8 (gnd) and 14 or 16 (vcc) start of with assuming you're daling with some kind of logic gate (Cmos). These are slightly more difficult. Determine if this IC has pins connected to each other. If so, work from the assumption that with logic gates the outputs cannot be connected to other outputs*. From there detemine if you can find circuit that suggests input or output for this IC and compare with the pinning of standard logic gates. Two adjacent pins connected? Expect them inputs and think of gates with multiple inputs (e.g. NANDS) See, if the expected output pin can function like an output..
For cases like this I use an older Elektor datasheetbook that just summs all CMOS and TTL gates, their pinning and their function. Very handy for quick comparison. Multiple pins other than 7 (or 8) to gound? Suggests grounded inputs as well.
* Exception to the rule is when can be absolutely certain that the outputs of the multiple gates always will have the same state. For instance because they're in parallel.
- Phasing of VCA stuff with power on pin 6 and 11 of a 14 pin IC? expect A LM13600 type chip.
- Partnumber partially sanded off? Use datasheetarchive and dump in there what you do have and see if any of the results could match your application. (That's how I traced this fcker in a George Dennis Wizzard Wah:http://www.datasheetarchive.com/pdf-datasheets/Datasheets-20/DSA-396453.pdf after part mining of the not completely sanded "L01")
- alternative housing shape, that's your first lead. Think of a MN3005 BBD for example. (does not fit? Maybe the sister MN3205 does.)
- Remember Soulsonics Klonclone tracing? A LT1044 type chargepump was expected and fit the pinning and circuit. Sand all you want, the circuit and measurements gave it's type away.
- Or MxrMxr's Tonebone classic degoop ?Out of the box thinking got him from the assumed "3 transistors" to the diode configuration that it was for real.

And therefore it all comes to knowing components and their pinning, as well as having quick access to overviews of component families. Details come later.

Hope this helps ye going ;)
Sorry. Plain out of planes.

http://www.dirk-hendrik.com
User avatar
Dirk_Hendrik
Old Solderhand
 
Posts: 4230
Joined: 03 Jul 2007, 09:44
Location: Old Amsterdam
Has thanked: 255 times
Have thanks: 1033 times

Re: Identifying IC chips that have been sanded down?

Postby RnFR » 06 May 2009, 15:32

man, great f'in post. tons of good shit in there, dirk. now this is what this place is about! i'm moving it to the FAQs section.
"You've converted me to Cubic thinking. Where do I sign up for the newsletter? I need to learn more about how I can break free from ONEism Death Math." - Soulsonic

Blog-APOCALYPSE AUDIO
User avatar
RnFR
Old Solderhand
 
Posts: 5138
Joined: 09 Jan 2008, 07:02
Location: Inner Earth
Has thanked: 150 times
Have thanks: 174 times

Re: Identifying IC chips that have been sanded down?

Postby Dirk_Hendrik » 06 May 2009, 15:59

Thanks ;)
Other than that... Experience. Just doing, doin, doin and doing again.

In the (at least for me) pre-internet age I spend about a day tracing my TS10, first time doing such a job...
Sorry. Plain out of planes.

http://www.dirk-hendrik.com
User avatar
Dirk_Hendrik
Old Solderhand
 
Posts: 4230
Joined: 03 Jul 2007, 09:44
Location: Old Amsterdam
Has thanked: 255 times
Have thanks: 1033 times


Return to ICs (Integrated Circuits)

Who is online

Users browsing this forum: No registered users and 3 guests