HASL for cartridges? (split/OT from: Cheese & Onion)

Modding and Technical Issues

Moderator: Moderators

norm8332
Vic 20 Enthusiast
Posts: 164
Joined: Sun Nov 13, 2016 11:04 am
Location: USA

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby norm8332 » Tue Dec 05, 2017 8:56 pm

I own a PLAnkton, like it very much and see where you are coming from concerning the voltage The design is sound. But as we know many people are using the original power supplies with the over voltage failure mode..In this case I'm afraid that the PLAnkton may suffer the same fate as some of the other chips. I think you could have achieved the desired 3.6V using an adjustable AMS1117-ADJ or similar then the PLAnkton would be protected from over voltage. You should consider it if there is a revision. It even could be added as a feature.

That said, I think people who are using the original supplies without power protection are not too concerned about blowing their C64 anyways.
Last edited by norm8332 on Tue Dec 05, 2017 9:26 pm, edited 1 time in total.

eslapion
ultimate expander
Posts: 4330
Joined: Fri Jun 23, 2006 7:50 pm
Location: Canada
Occupation: 8bit addict

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby eslapion » Tue Dec 05, 2017 9:23 pm

norm8332 wrote:I own a PLAnkton and like it very much and see where you are coming from concerning the voltage. But as we know many people are using the original power supplies with the over voltage failure mode..In this case I'm afraid that the PLAnkton may suffer the same fate as some of the other chips. I think you could have achieved your desired 3.6V using an adjustable AMS1117-ADJ or similar then the PLAnkton would be protected from over voltage. You should consider it if there is a revision.

That said, I think people who are using the original supplies without power protection are not too concerned about blowing their C64 anyways.

A few PLAnktons were inserted by accident in the SID socket and were therefore exposed to a source of 12Vdc and survived the 'ordeal'. There are protections inside the XC9536XL which don't exist in Commodore original circuits and the presence of the resistor adds to these protections.

The SID uses both 12Vdc and 5Vdc and it so happens the 12Vdc (9Vdc for the 8580) is fed on pin 28, right where the PLA expects 5Vdc. A real PLA would have busted.

Honestly, I don't know what additional protection I could put while keeping PLAnkton at the most affordable price.

The AMS1117-ADJ (which I couldn't find on Digikey) costs more than 1$ and requires multiple external components.

This being said, this is clearly an attempt by MCes at causing a diversion. Since he can't admit it's a bad idea to sell cartridges which have HASL on the edge connector, he's trying to nitpick on a network of components which operate in a way he doesn't understand and assume it's some sloppy job and he assumes the people at large will see it that way too. Once more, bad news, there's far more than meets the eye.

He would've noticed it immediately if he had taken measurements with a multimeter or tried to simulate the circuit using Microcap or PSpice.

The usage of a combination of a diode and resistor to create a non-linear voltage divider can be made more clear here: https://hackaday.io/project/2147-7wave- ... ng-circuit
No one can do everything. Everyone can do something.

norm8332
Vic 20 Enthusiast
Posts: 164
Joined: Sun Nov 13, 2016 11:04 am
Location: USA

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby norm8332 » Tue Dec 05, 2017 9:35 pm

Haha you got to it before I edited it...Yeah that's good to know that it survived the 12V. Like I said I have no issues with the way you did it...Of course the Chinese AMS1117-ADJ can be had for 5¢ each shipped, But I know you like to use only genuine parts.

eslapion
ultimate expander
Posts: 4330
Joined: Fri Jun 23, 2006 7:50 pm
Location: Canada
Occupation: 8bit addict

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby eslapion » Tue Dec 05, 2017 9:46 pm

norm8332 wrote:Haha you got to it before I edited it...Yeah that's good to know that it survived the 12V. Like I said I have no issues with the way you did it...Of course the Chinese AMS1117-ADJ can be had for 5¢ each shipped, But I know you like to use only genuine parts.

The MCP1700T3302 regulator which I use for TOLB is 0.43$ but the case of TOLB is different.

TOLB has components which MUST be fed 3.3V within 5% tolerance to operate properly and the 2 outputs are directly fed to the VIC-II which, unlike 74HC logic chips have a logic threshold of 1.3V. In this case, using less than 3.5V is perfectly fine.
No one can do everything. Everyone can do something.

norm8332
Vic 20 Enthusiast
Posts: 164
Joined: Sun Nov 13, 2016 11:04 am
Location: USA

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby norm8332 » Tue Dec 05, 2017 9:49 pm

One more off topic...Digikey just doesnt carry the AMS brand but they have them. You just search 1117 and select voltage regulators then adjustable in the filters.

eslapion
ultimate expander
Posts: 4330
Joined: Fri Jun 23, 2006 7:50 pm
Location: Canada
Occupation: 8bit addict

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby eslapion » Tue Dec 05, 2017 10:28 pm

norm8332 wrote:One more off topic...Digikey just doesnt carry the AMS brand but they have them. You just search 1117 and select voltage regulators then adjustable in the filters.

Yes, I got them. But honestly, the objective of an electronics designer is to get the least expensive and simplest solution which achieves the job reliably with a good factor of resilience.

So far I have had zero return or defect caused by a problem with the power source. If some guy looks at it from the outside without even noticing it's not giving the standard voltage then I guess it's a compliment.

Besides, the least expensive regulator in the family you suggested is 0.59$ while the actual components used on PLAnkton for power regulation is 0.15$ or a quarter the price for the same result. Are we really going there ?
No one can do everything. Everyone can do something.

norm8332
Vic 20 Enthusiast
Posts: 164
Joined: Sun Nov 13, 2016 11:04 am
Location: USA

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby norm8332 » Tue Dec 05, 2017 10:46 pm

No we aren't going there...As I said I don't have an issue with the design. I was just pointing out alternate options. And as for the HASL thing, I think for now we can agree to disagree (about permanent damage to the VIC's connector) and leave it at that. I do agree that the cartridge itself will benefit from gold contacts.

User avatar
highinfidelity
Vic 20 Afficionado
Posts: 407
Joined: Thu Jul 28, 2011 2:34 am
Website: http://www.hirtel.it
Location: Torino, Italy.

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby highinfidelity » Wed Dec 06, 2017 2:40 am

eslapion wrote:I suppose Mike is an integralist too then... :lol:

All the contrary, Mike has written on this matter generic comments that we all agree with: gold palted connectors would work better, etc. I'll add that platinum-iridium plated connectors would work even better. Wow. If I was chief engineer at the International Space Station, I would definitely require that all comb connectors would be platinum-iridium plated, and would reject everything else. Cool. 8)

Now back to Earth. What we have here is a home computer that - perhaps excepting the keyboard - was entirely designed, engineered and produced with what is know as the "good enough" industrial principle in mind.

Mike, I (we) agree with you that a gold plated connector is the state of the art, however have you ever had one of your VIC connectors irreparably damaged by a HASL connector, and only by that, that is you have the absolute certainty that the HASL connector and the HASL connector alone was the cause of the damage, excluding then all other possible causes (wrong mechanical dimensions of the PCB, unaccurate thickness of the male connector, approximate finishes, wrong protective coatings, ambient polluting compounds, humidity, mold and alike...), or do you personally know (name and surnames, then excluding all internet fake news and reports, post-truths and alike) people who has had their VIC connector irreparably damaged by a HASL connector, and which connector you have personally inspected and tested, deducing with absolute certainty that no other damage cause than the insertion of a HASL male connector alone was the cause of such damage?
GOD is REAL. Unless declared DOUBLE PRECISION.

eslapion
ultimate expander
Posts: 4330
Joined: Fri Jun 23, 2006 7:50 pm
Location: Canada
Occupation: 8bit addict

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby eslapion » Wed Dec 06, 2017 4:19 am

highinfidelity wrote:... or do you personally know (name and surnames, then excluding all internet fake news and reports, post-truths and alike) people who has had their VIC connector irreparably damaged by a HASL connector, and which connector you have personally inspected and tested, deducing with absolute certainty that no other damage cause than the insertion of a HASL male connector alone was the cause of such damage?

As said above, the last slot of my Cardco Cardbus 3 ... owned by myself and replaced in 2015... But you try to make it specific to the VIC-20, it's not.

Image
That's the Behr-Bonz clone mentioned a few months ago by MCes. Rather obvious...
Higher resolution here: https://c1.staticflickr.com/3/2645/4148 ... d889_b.jpg

Image

Image
I find it insulting the people who made these clones put "By Eslapion & Viande" on the top silk screen...

The real one...
Image
BTW, Since you can see B.C. handwritten on the sticker instead of PAL or NTSC, this is a custom made version so I know exactly who bought it from me.

All the contrary, Mike has written on this matter generic comments that we all agree with: gold palted connectors would work better, etc. I'll add that platinum-iridium plated connectors would work even better. Wow. If I was chief engineer at the International Space Station, I would definitely require that all comb connectors would be platinum-iridium plated, and would reject everything else.

That's one incredibly clunky comparison. HASL is not designed AT ALL to handle the strong friction which occurs when you insert a board in a female edge connector. It wears out instantly. Mike posted a link to an article which says exactly that and I think he did so because he believes it too.

Gold plating is not a better option (or precious metal plating since not only gold does the job), it is the absolute minimum required. I tested HASL boards on my own equipment knowing the risks. MCES sells memory expansions with HASL edge connectors to people who don't know about that without telling them ANYTHING.

Even worst, he defends what he does with tooth and nail... and you seem to support this.
Last edited by eslapion on Wed Dec 06, 2017 2:00 pm, edited 1 time in total.
No one can do everything. Everyone can do something.

User avatar
MCes
Vic 20 Hobbyist
Posts: 142
Joined: Fri Jul 24, 2015 1:19 am
Location: Italy

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby MCes » Wed Dec 06, 2017 9:36 am

Mmmm... comparing...
"MCES sells memory expansions with HASL edge connectors to people who don't know about that without telling them ANYTHING.
Even worst, he defends what he does with tooth and nail...
"
but I wrote:"Hasl, gold, hard gold, Nothing of this are perfect... "
and you can see that is not a problem to inform people: http://sleepingelephant.com/ipw-web/bulletin/bb/viewtopic.php?f=3&t=8582#p95263

MCes: "For example a movible connections done with HARD GOLD could be dangerous: the hardest surface scratch the softest surface, so the hard gold pcb could scratch the metal part of connection inside the VIC20 connector...... do you want this?"
it's a doubt, not only mine:
Eslapion wrote: " HASL is not designed AT ALL to handle the strong friction which occurs when you insert a board in a female edge connector."
The differences is that I'm focusing on don't scratch the VIC20 connector, Eslapion is focused on don't scratching it's side of connections (the cartridge)!

Eslapion wrote: "But honestly, the objective of an electronics designer is to get the least expensive and simplest solution which achieves the job reliably with a GOOD FACTOR of resilience.",

GOOD FACTOR isn't an "exact science", it's a good choice of the designer, but only if designer is Eslapion.

Choice of the designer......
Eslapion wrote:"To ensure compatibility with the C64 Reloaded MK1 which has a few 74HCxx logic circuits as well as all C64 which may have been repaired with this type of chips, the XC9536XL on PLAnkton operates at 3.6V(..)The recommended operating voltage is from 3.0 to 3.6"
"The voltage drop of the resistor and diode is 0.022mAx27Ohms+0.75v=1.344v. Then 5V-1.344=3.656, a tad above the recommended 3.6v but well below the absolute maximum rating of 4.0v."


From XC9536XL datas: "Exposure to Absolute Maximum Ratings conditions [3,6V<Vcc<4.0V] for extended periods of time may affect device reliability." (and I don't inform people about danger from my cartridge...)

A good work!

if the "level critical" output was configurated "open drain" with external 5V pull-up all could work at proper level, you can find this method and another one that speed up the rising edge of pull-up here: https://www.xilinx.com/support/document ... /ug445.pdf

Choice of the designer:
-XC9536XL (3,3Vcc, 5Vtolerant) instead XC9536 (5Vcc, HC compatible): maybe 1$ difference
-XC9536XL without external resistor for 5V pull-up on a pair of output (it could be 3,3Vcc and 5Voh) because 2 resistor are too much expensive!

Ok(?), so you CHOSE to use this particular CPLD without resistors, it can be done but with EXACTLY Vcc=3,6V whit a PLA Vcc that is 5V+/- 5% (like as every chips inside C64, PLA too).

+/-5% on 5V is a variability of 0,5V that can't be imported on 3,6V Vccpin of CPLD, and we have to sum that the drop trought a resistor-diode is not fix, it can vary with:
TOLLERANCE of Vf (+-20%...25% )
TEMPERTURE (on diode)
PLA activity = current needed (dorp on resistor)

Who buy this have only to hope that Xilinx engineers are a lot conservative in their chips design.....

Do you never informed the PLAnkton buyers on what they was buying?
Eslapion want apply at other people the law that don't want apply to himself.
“Don’t argue with idiots because they will drag you down to their level and then beat you with experience” (Mark Twain ?)

eslapion
ultimate expander
Posts: 4330
Joined: Fri Jun 23, 2006 7:50 pm
Location: Canada
Occupation: 8bit addict

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby eslapion » Wed Dec 06, 2017 12:07 pm

MCes wrote:From XC9536XL datas: "Exposure to Absolute Maximum Ratings conditions [3,6V<Vcc<4.0V] for extended periods of time may affect device reliability." (and I don't inform people about danger from my cartridge...)

A good work!

No problem buddy! The datasheet says the AMR is 4.0V or above, not 3,6V<Vcc<4.0V...

PLAnkton and the other modules I sell are all guaranteed for 3 years. I fully honor this guarantee even when the problem isn't my fault (damaged pins...). I have sold more than 1100 of them so far and I didn't see a flurry of replacement requests because the CPLD has busted, in fact I never had any.

if the "level critical" output was configurated "open drain" with external 5V pull-up all could work at proper level, you can find this method and another one that speed up the rising edge of pull-up here: https://www.xilinx.com/support/document ... /ug445.pdf

Nope. The C64 uses NMOS and TTL-LS technology chips. This configuration is recommended for a CMOS IC powered at 3.3V signaling to CMOS technology powered at 5V... I configured PLAnkton to work with a C64 that uses Commodore ICs. For these, the proper level isn't the same at all...

See: https://www.allaboutcircuits.com/textbo ... ge-levels/

But I'll let you in on a little secret... this is the equivalent circuit of a TTL NAND gate. https://image.slidesharecdn.com/myslide ... 1411190631
What do you see on the input ? Yup... every logic chip in a VIC-20, C64 and 1541 drive has pull-up resistors built right into their inputs. Why bother to add any ?

Choice of the designer:
-XC9536XL (3,3Vcc, 5Vtolerant) instead XC9536 (5Vcc, HC compatible): maybe 1$ difference
-XC9536XL without external resistor for 5V pull-up on a pair of output (it could be 3,3Vcc and 5Voh) because 2 resistor are too much expensive!

Ok(?), so you CHOSE to use this particular CPLD without resistors, it can be done but with EXACTLY Vcc=3,6V whit a PLA Vcc that is 5V+/- 5% (like as every chips inside C64, PLA too).

Nope. It's because the XC9536 is DISCONTINUED!! I don't use surplus or counterfeit parts... or old ones which contain lead. All parts on PLAnkton are genuine and RoHS compliant ONLY! (BTW, why exactly did you cover in black ink the markings on the RAM chip of your 32k expander ??)

If you signal at 3.5V or more, then pull-up resistors are not required for 5V powered HC compatibility. PLAnkton signals at about 3.6V so... no problem there. There are EIGHT outputs on PLAnkton (or a genuine PLA), not 2 so EIGHT pull-up resistors would be required according to you.

+/-5% on 5V is a variability of 0,5V that can't be imported on 3,6V Vccpin of CPLD, and we have to sum that the drop trought a resistor-diode is not fix, it can vary with:
TOLLERANCE of Vf (+-20%...25% )
TEMPERTURE (on diode)
PLA activity = current needed (dorp on resistor)

Who buy this have only to hope that Xilinx engineers are a lot conservative in their chips design.....

Do you never informed the PLAnkton buyers on what they was buying?
Eslapion want apply at other people the law that don't want apply to himself.

You don't account for the exponential conduction curve of the diode... this is a combination of LTI and non-linear components which include the diode and the power consumption of the CPLD.

You also don't account for the fact the XC9536XL on PLAnkton is tied to a 100pF capacitor which it charges and discharges about 2.02 million times per second. Since the energy stored in a capacitor is calculated 1/2xCxV^2, this means the energy expended increases to the square of the voltage. If the voltage rises, the system will consume more power and stabilizes itself.

... you are welcome.
No one can do everything. Everyone can do something.

User avatar
MCes
Vic 20 Hobbyist
Posts: 142
Joined: Fri Jul 24, 2015 1:19 am
Location: Italy

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby MCes » Thu Dec 07, 2017 8:45 am

Eslapion wrote:
"The datasheet says the AMR is 4.0V or above, not 3,6V<Vcc<4.0V..."
I have problem with english language, but you (if you are honest) have oculistic problem: Vcc AMR is from -0.5V to 4.0V at exlusion of Recommended Operation Conditions (ROC) from 3.0V to 3.6V, so if 3,6<Vcc<4.0 then Vcc is inside AMR and out of ROC.
You try to cnvince thet Vcc= -0,3V could be a good project choise!

I suggest to read this short article, it's clear and the author appear reliable.http://www.analog.com/en/analog-dialogue/raqs/raq-issue-50.html

Eslapion:
I have sold more than 1100 of them so far and I didn't see a flurry of replacement requests because the CPLD has busted, in fact I never had any.

This is the proof that Xilinx engineers (at opposite as you)had a good conservative designe choose, not else.

Eslaption wrote :
The C64 uses NMOS and TTL-LS technology chips. CMOS IC powered at 3.3V signaling to CMOS technology powered at 5V ... I configured PLAnkton to work with a C64 that uses Commodore ICs. For these, the proper level is not the same ..."
Into C64 the address bus A12,13,14,15 are 5V pull-upped by 3,3K to determinate 1111xxxx.xxxxxxx when VICII get BUS and put CPU in 3state,
tecnically the same solution that I suggest and Xilinx suggest for XC9536XL (3,3Vcc) that produce a 5V rail to rail output!

Eslaption think that everybody else are wrong: I (and this could be..) , Xilinx engineers, and now Commodore engineers developers too!

Eslapion:
There are EIGHT outputs on PLAnkton (or a genuine PLA), not 2 so EIGHT pull-up resistors would be required according to you.

OK: you has admitting that your (bad) choice is motivated by cost of 8 resistors, not 2.
Anybody can see the cost of the just choose (9pin net resistors): less than 0.1$!!!!!!

Eslapion:
You don't account for the exponential conduction curve of the diode... this is a combination of LTI and non-linear components which include the diode and the power consumption of the CPLD.
You also don't account for the fact the XC9536XL on PLAnkton is tied to a 100pF capacitor which it charges and discharges about 2.02 million times per second. Since the energy stored in a capacitor is calculated 1/2xCxV^2, this means the energy expended increases to the square of the voltage. If the voltage rises, the system will consume more power and stabilizes itself.


I never laughed so much before! :lol: :P :D :lol:
For you the gaussian distribution of production tolerance of Vf is a vertical line!
For you the reducing of drop voltage of diode (Vf) with rising of temperature is a tale for child!
For you into a C64 the Vcc between pin14 and pin18 of PLA can be only 5.000000000V without noise overlapped!

The discussion about energy stored in a capacitor is RIDICULOUS, only sand into the eyes.

You are offending the mind of anybody who is reading!

I think that we were side by side when somebody ask us "Do you want the blu or the red pill?"
Who is reading can made a own idea about who is living in the real world and who not.
Last edited by MCes on Thu Dec 07, 2017 9:59 am, edited 1 time in total.
“Don’t argue with idiots because they will drag you down to their level and then beat you with experience” (Mark Twain ?)

User avatar
MCes
Vic 20 Hobbyist
Posts: 142
Joined: Fri Jul 24, 2015 1:19 am
Location: Italy

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby MCes » Thu Dec 07, 2017 9:53 am

How could be if MCes and Xilinx tips was (will be) adopted?
a PLANkton "next generation": 2x 4resistors SMD network and the 3.3V voltage regulator on opposite side...

PLANkton next.JPG


The SMD network could be in back side, for simplify the PCB redrawing....

Hey FORUM, what do you think about?
(I want popcorn to!)
Last edited by MCes on Thu Dec 07, 2017 10:15 am, edited 1 time in total.
“Don’t argue with idiots because they will drag you down to their level and then beat you with experience” (Mark Twain ?)

eslapion
ultimate expander
Posts: 4330
Joined: Fri Jun 23, 2006 7:50 pm
Location: Canada
Occupation: 8bit addict

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby eslapion » Thu Dec 07, 2017 10:04 am

These resistors aren't necessary buddy. All ICs used in the C64 are designed for TTL level signals. These are below 0.8 V for a logic 0 and above 2.7V for a logic 1.

PLAnkton is designed to signal 0-3.6V so... according to you, even a genuine Commodore PLA would require these resistors because they too signal at a similar voltage range. Total nonsense...

The XC9536XL Absolute Maximum rating is 4.0V.

For you into a C64 the Vcc between pin14 and pin18 of PLA can be only 5.000000000V without noise overlapped!

Well, why don't you actually power a PLAnkton with 5.5V (in a running C64) and check the voltage at the CPLD and see what happens...
Last edited by eslapion on Thu Dec 07, 2017 10:10 am, edited 1 time in total.
No one can do everything. Everyone can do something.

User avatar
MCes
Vic 20 Hobbyist
Posts: 142
Joined: Fri Jul 24, 2015 1:19 am
Location: Italy

Re: HASL for cartridges? (split/OT from: Cheese & Onion)

Postby MCes » Thu Dec 07, 2017 10:08 am

Repeat a FAKE never made it TRUE


I think that our positions are not conciliably.
In absence of other (serious/credible) material I'll wait the opinion of other people.
“Don’t argue with idiots because they will drag you down to their level and then beat you with experience” (Mark Twain ?)


Return to “Hardware and Tech”

Who is online

Users browsing this forum: No registered users and 1 guest