Retro Chip Tester Pro

[cliff_poole]

I just bought cheap unbranded ones from eBay. Had to buy a second batch later to get enough good ones, and was still one short. Luckily, I found one in my tray of odd zeners, it even looked close enough to the others, saved me buying a third batch.

I even considered using the slightly out of spec ones, but then remembered how often I've come across faulty 4116s shorting supply lines in Spectrums. Usually takes out the ZTX650 transistor in a Speccy, which annoyingly, there are virtually no equivalents for.

 

[TheDaddy]

Some on Ali but christ knows if they are good or not !

100pcs 1w 5.1v 1n4733a 5v1 1n4733 Do-41 Zener Diode - Diodes - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

vi.aliexpress.com

Dave.

 

[Phils Arcade]

When getting back into electronics, I decided to use Ali for bulk resistor packs, as in, how bad can they be, they're resistors.
Some were good, others were so far out, felt like they belonged in a different value bin altogether. Ended up going back to RS and CPC Farnell for them. Cricklewood Electronics isn't bad either.

 

[TheDaddy]

When getting back into electronics, I decided to use Ali for bulk resistor packs, as in, how bad can they be, they're resistors.
Some were good, others were so far out, felt like they belonged in a different value bin altogether. Ended up going back to RS and CPC Farnell for them. Cricklewood Electronics isn't bad either.

Yeh its a 50/50 with Ali stuff. I use caps and all seem ok but am aware that voltages can be way out so I just use bigger voltage ones. I dont buy from ebay as people just buy of Ali and sell on ebay or its the people off Ali selling on there at 4 times the price. Although ali's prices have gone up so much over the years.

As you say phil the only real way is RS really, Its my goto when I need reliable stuff (Not that often now adays I as use as many spares from otehr PCB's as poss)

Dave.

 

[Macro]

First batch came from Digikey, but I've ordered a couple of bundles from ebay (UK based) - different manufacturers, so fingers crossed.

 

[Georgian2]

The more fake china shit you buy, the more crap they'll sell us. Just stop buying from them and pay a little more for genuine stuff. We are encouraging them to so.

 

[Macro]

But the genuine stuff is made in China anyway...

 

[cliff_poole]

And in this case, the diodes are all within their stated tolerances, so we are not having a moan at the sellers or anything.

The RCT just requires a very specific tight tolerance which aren't available to buy. You have to buy standard ones and grade them yourself. Not a problem, they don't cost a great deal.
It is important to get right as they are there to protect the I/O of the microcontroller against over or under voltage. Some of the devices under test might use -5V and 12V so this is a risk if the chip you are testing is faulty in a weird way.

 

[Georgian2]

The 1n4733 is still easly available. Am I missing something here?

 

[cliff_poole]

The 1n4733 is still easly available. Am I missing something here?

Yep. Even the Vishay branded ones have a +/- 10% zener voltage tolerance which is why you have to test them all, and that's why I went for the cheap ones.

 

[Macro]

I ordered 100 from ali as well, only cost £3 !!

hopefully I can find 32 that meet the criteria out of the 200 I have on the way, if I do I doubt I will ever buy a 5.1v zener again

 

[Macro]

Well latest batch of diodes arrived.

37 passed the 9v test
on the 5v test, 10 rated as good (4.85v +), 13 as OK (my own rating 4.75v +) and 14 as acceptable (4.70v +)

so in theory I now have the 32 needed, but since I have more on the way, I'll see if I get enough "good" and "OK" to just use those.

Big improvement on the previous batch 0 usable from 64 (although I guess they were consistent!)

 

[cliff_poole]

Yesterday I was looking at a bootleg Pac Man, and removed two of the 7489 RAMs for testing. Both consistently failed tests in the RCT.

As I have the optional DCDC board that allows increasing the +5V, I decided to try it for the first time. With the test voltage set to 5.15V, one still failed, but the other would occasionally pass. At 5.3V, (the highest it can be set), both would pass most times I hit the test button, but occasionally fail.

I replaced both chips anyway as they are definitely faulty, but it satisfied my curiosity.

 

[Ozmo]

I just made mine up. I bought the boards without mega chip.
I got most of the parts from the BOM supplied from the German site included (30+10 postage). Cheap and easy. Pin headers+jumpers excluded as they were only thing overpriced.
The screen came from Amazon(10 euros).
Mega chip desoldered from an unused arduino mega clone. Programming the mega requires an expensive programmer i was lucky to have access to- id recommend choosing option to getting that on board already soldered and programmed.

However i forgot to test the zeners…doh… logic chips test perfectly- but all my dram and s ram chip fails. Dont know if the chips or the tester:/

Can the zeners be tested now without desoldering them anyone know?? They are in so neat lol.

 

[Macro]

it depends on your luck, since at some point you have to connect them to a 9v supply / battery then you could be sending too many volts to the mega chip. (although I tested 250 diodes and none were too high, most were too low)

 

[Ozmo]

Found one of the issues anyway - to measure my Drams (D41464C) - (I test them as a 4464 chip as required) - but you need add a jumper on J6 where it says "4416/64"
Now all 10 Drams succeed at the first few tests and then all fail at the same this exact same point - so I think the hardware is correct - wondering what I could be doing wrong...?

----------------
Testing 4464
Checking: (tests 0, 5, A, F ok) count+ (fails)
Features: PM
[][][][][][][][][][]
----------------
4464 Test Failed
A:0 0>8
----------------

I flashed the provided diagnostics hex file - it lets you manually set all the pins to VCC.
Using USB or 7.5V via barrel jack - Using a multimeter - I read all the Pins set to VCC the exact same at 4.60Volts
(its not 5V as this drop is due to the 5.1 zeners the 5v goes through - you think 4.6v is high enough?)

I have tried firmware upload_1.0_pro_v0.24 and upload_1.0_pro_v0.26

Any ideas why all 10 drams are failing anyone? They were just arrived - Could be all dodgy, but would be unusual they all pass most of the tests and fail the exact same way...?
ta.

 

[Macro]

4.6 is too low from the docs. 4.7 is marked as usable, but higher is better. I got 32 at 4.8 or higher from testing 264 (in 3 batches from different places) The ones I got from digikey were all too low!

n.b. the PDF says ...
"When you measure the Zener diode, a value of 5.0V is very good, a value of at least 4.85V is good, a value of at least 4.7V is just about acceptable."

 

[cliff_poole]

Any ideas why all 10 drams are failing anyone? They were just arrived - Could be all dodgy, but would be unusual they all pass most of the tests and fail the exact same way...?
ta.

I guess it could be a power issue. The only way I can think to test that theory would be to make the optional DC\DC board that can be jumpered to a higher supply voltage.

The decoupling cap jumper is important on older devices because the tester applies power and then immediately starts the test. Some older chips, especially 4116s, draw more current, so can cause a dip in voltage as the test starts.

It might be worth asking on the RCT forum. Someone there is bound to know exactly what that test is doing. Maybe the test is incompatible with the chip type, or a setup option needs to be tweaked ?, (there are options for pullup resistors, although I think that is just for open collector output devices).

 

[Ozmo]

I guess it could be a power issue. The only way I can think to test that theory would be to make the optional DC\DC board that can be jumpered to a higher supply voltage.

The decoupling cap jumper is important on older devices because the tester applies power and then immediately starts the test. Some older chips, especially 4116s, draw more current, so can cause a dip in voltage as the test starts.

It might be worth asking on the RCT forum. Someone there is bound to know exactly what that test is doing. Maybe the test is incompatible with the chip type, or a setup option needs to be tweaked ?, (there are options for pullup resistors, although I think that is just for open collector output devices).

I figured ut out! I had the decoupling jumper ON - (to aid soldering it in straight and forgot remove it.)
When jumper is off - all tests pass perfectly. Manual says leaving in should be ok- but definitely it fails consistently.
its passing the definitely broken ram chips I took from the spectrum though :S…

Spectrum +3 working well now too

Thanks all!

 

[cliff_poole]

Excellent !.

Love the +3, it was my last Spectrum before going over to an Amiga. Also my first Multiface. It was amazing to be able to save my position in Head Over Heels to disk, without having to start from scratch every time.