UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
 
Thread Tools
Old 9th Apr 2021, 3:00 am   #161
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: The Transam Triton Personal Computer

It looks like the chips go face down on the ceramic substrate. Maybe an early version of a bga. Is there an insulating layer on the circuit side of the chip with round holes in the insulating layer for solder balls?

You might check if the remains on the ceramic will melt with a soldering iron or hot air pencil, just to check if it was an adhesive instead.

Then a small amount of paste on each chip pad. Reflow with hot air pencil to form a ball. Line up on the pads and reflow with hot air pencil after applying plenty of flux.
Mark1960 is online now  
Old 9th Apr 2021, 9:47 am   #162
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by Mark1960 View Post
It looks like the chips go face down on the ceramic substrate. Maybe an early version of a bga. Is there an insulating layer on the circuit side of the chip with round holes in the insulating layer for solder balls?

You might check if the remains on the ceramic will melt with a soldering iron or hot air pencil, just to check if it was an adhesive instead.

Then a small amount of paste on each chip pad. Reflow with hot air pencil to form a ball. Line up on the pads and reflow with hot air pencil after applying plenty of flux.
Well I blasted the ceramic substrate with a hot air solder station at 400c then used a soldering iron then wiped but the ball still remains.
Attached Thumbnails
Click image for larger version

Name:	SingleShot0001.jpg
Views:	77
Size:	77.0 KB
ID:	231252  
GeraldSommariva is offline  
Old 9th Apr 2021, 11:36 am   #163
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by GeraldSommariva View Post
Quote:
Originally Posted by Mark1960 View Post
It looks like the chips go face down on the ceramic substrate. Maybe an early version of a bga. Is there an insulating layer on the circuit side of the chip with round holes in the insulating layer for solder balls?
>>
>>
Well I blasted the ceramic substrate with a hot air solder station at 400c then used a soldering iron then wiped but the ball still remains.

Yes, having a closer comparison between marks left behind on the substrate (and that two pads are also slightly-closer on one side, as well as the contact-marks), then the chip must have been fitted "upside-down", similar to more recent Chip Scale Packaging of the die being mounted directly on the PCB (Similar to BGA, but that often has a very small thin carrier PCB that the die is mounted to / often plastic over-moulded with part numbering on top, rather than just the raw die)
- It had been difficult to tell on a fitted-one, because all the covering adhesive had masked clear visibility of the surface

I did wonder what would have prevented shorts to the rest of the surface of the chip, but it probably has some clear passivation layers over that. And also, I note from this photo that there was also clear adhesive over the surface, with pockets around the pads, as can be seen on this photo: https://sites.google.com/view/transa...ensor-module-5
However, rather than the chip originally being covered with protective clear adhesive, it may well have flowed underneath it when the the top surface of the substrate had been covered with the protective clear dome of this.

It was difficult to tell from the substrate photos, whether the marks on the pads were actually raised above the surface. But if so, then most likely it was actually silver-loaded conductive epoxy, that has sheared away from the pads on the die.
And,given that the whole of substrate's pads are not covered with the adhesive, it does look like there was originally small balls of it applied directly to the chips surface, then accurately positioned on the substrates - probably all done semi-automatically as would be difficult to get the consistency by hand.

It does seem most likely that the chip was originally mounted with this conductive-epoxy, as I have seen accelerometer sensors that used this - Although they were ceramic packaged , and just had 2 rectangular gold pads. The originally had recommended much more exotic Gold epoxy, but silver could also be used. However, later versions just used solder-paste to make it much easier to mount with the other Surface Mount auto-assembling.

If you can get solder paste to form solder-balls on the chip's pads, by applying and heating with hot air pencil, then it may be possible to flip it over and re-heat from other side to reflow it onto the substrate (after applying solder paste to that). However, heating the ceramic substrate enough might be tricky, without a hot-plate designed for this.
It may be possible to do it in one of those DIY reflow ovens, people makes by converting a £30 "Pizza" mini-oven.
ortek_service is offline  
Old 10th Apr 2021, 9:57 am   #164
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by ortek_service View Post


It was difficult to tell from the substrate photos, whether the marks on the pads were actually raised above the surface. But if so, then most likely it was actually silver-loaded conductive epoxy, that has sheared away from the pads on the die.
And,given that the whole of substrate's pads are not covered with the adhesive, it does look like there was originally small balls of it applied directly to the chips surface, then accurately positioned on the substrates - probably all done semi-automatically as would be difficult to get the consistency by hand.
The pads are raised above the surface you can just feel them with your finger tips.

Quote:
Originally Posted by ortek_service View Post

If you can get solder paste to form solder-balls on the chip's pads, by applying and heating then it may be possible to flip it over and re-heat from other side to reflow it onto the substrate (after applying solder paste to that). However, heating the ceramic substrate enough might be tricky, without a hot-plate designed for this.
It may be possible to do it in one of those DIY reflow ovens, people makes by converting a £30 "Pizza" mini-oven.
like the sound of the DIY reflow ovens, I did want to have ago at using surface mount chip's on home made PCB as many components are not available in DIP form also it saves on all that hole drilling. if anyone as done this tip's would be good.

I will try an re-solder the hall sensor with my hot air pencil, the solder past that I have is oldish and there is no "expiration date" on it. does solder past expire? Is it still good?
GeraldSommariva is offline  
Old 10th Apr 2021, 6:52 pm   #165
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: The Transam Triton Personal Computer

You can apparently obtain solder in small uniform ball sizes, that are used to re-ball BGA's from unleaded to leaded for people exempt from using lead-free, for best reliability. But may not be easily available to those not doing the re-balling process, that actually often seemed to result in damaged IC's.

Practical Electronics magazine recently had a project to make a SMD-Reflow unit based on a "Pizza" oven. And no doubt Elektor and details online many years ago. I do know a few people who've done this, with one at least just using it as is, with a thermocouple and manually stepping temperature up & down. But I think others had made their own controller boards.


I've had some solder paste in the fridge for > 20years, since I bought a surplus unopened pot. But never got round to trying it.

The only time I used paste was in a syringe, with twist-on fine metal nozzle > 25yrs ago at work, to build up solder on pads on a alumina substrate as trying to use a soldering iron on it when not pre-heated to 150degC etc. could result in the gold leeching away, as it wasn't palladium etc. plated.
I know you could get compressed air attachments to dispense controlled amounts of solder from these, for production.

Silver-loaded epoxy usually has a quite short shelf life, of a few months
- Even when stored in a fridge. And often not cheap from RS etc
ortek_service is offline  
Old 3rd May 2021, 10:43 pm   #166
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: The Transam Triton Personal Computer

Gerald has not posted for a while but, he has updated his website with the first scan of the Independent User Group magazine so he is making progress.

I also got around to documenting my progress (and that of some others) to date... https://youtu.be/xn2DQiin3EQ
Timbucus is offline  
Old 7th May 2021, 10:14 am   #167
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: The Transam Triton Personal Computer

If anyone is building a Triton then there is an offer here for the needed VDU chip....

https://www.vintage-radio.net/forum/...&postcount=905
Timbucus is offline  
Old 7th May 2021, 10:52 am   #168
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: The Transam Triton Personal Computer

Can you maybe rustle up interest on the Triton groups you are involved with? I don't think it's likely that I will ever build the PE VDU after all the work that Karen, bless her, put into OrtonView - it's just so much easier for anyone who wants an MK14 VDU to build the latter.
SiriusHardware is offline  
Old 11th May 2021, 1:21 pm   #169
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by Timbucus View Post
Gerald has not posted for a while but, he has updated his website with the first scan of the Independent User Group magazine so he is making progress.
Yes been out of action I had a bit of a mishap and broke my left arm (humerus bone near the top) having the use of one arm has made progress on my Triton project slow. Being left handed its been challenging using hand tools and soldering is near impossible.

I have updated my site with another Independent User Group magazine ( No.2 Feb 1983) Transam-Triton

Also have made some progress with my Triton keyboard which is proving to be a bit of a mission.
GeraldSommariva is offline  
Old 11th May 2021, 5:57 pm   #170
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: The Transam Triton Personal Computer

Oh sorry to hear that Gerald nasty - thanks for sharing another magazine despite the injury - I will have a read.

Good luck with the keyboard I am sure it will be worth it in the end.
Timbucus is offline  
Old 13th May 2021, 8:41 pm   #171
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by ortek_service View Post
You can apparently obtain solder in small uniform ball sizes, that are used to re-ball BGA's from unleaded to leaded for people exempt from using lead-free, for best reliability. But may not be easily available to those not doing the re-balling process, that actually often seemed to result in damaged IC's.
I have now given up on trying to re-solder the keyboard sensors IC back onto the ceramic substrate, after many attempts non would work.

By chance I found a company in the Czech Republic that will ship to the UK and they have the TESLA Hall sensors, DSM Components

So this MicroSwitch Keyboard might just make it back to working condition.
GeraldSommariva is offline  
Old 13th May 2021, 9:37 pm   #172
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: The Transam Triton Personal Computer

I have ordered from DSM in the recent past, cassette tape heads. It was a bit slow as the order was lost in their old system, but they sorted it out quite quickly when I contacted them by email. Originally saw them on ebay, but their own website was cheaper.
Mark1960 is online now  
Old 13th May 2021, 10:57 pm   #173
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by Mark1960 View Post
I have ordered from DSM in the recent past, cassette tape heads. It was a bit slow as the order was lost in their old system, but they sorted it out quite quickly when I contacted them by email. Originally saw them on ebay, but their own website was cheaper.
Yes I found them on ebay, then ended up on their web site, the shipment time are long 3 days to 2 weeks, in EU, 1-4 Weeks for Rest of World.

I whatsapp's them regarding using a carrier like DHL, they recommended sticking with Česká pošta with tracking due to Brexit Complications so my parts should turn up next week.

They also had the video memory chips I needed for the Trition (2102/2, 1K Static Ram) picked up 4 for £2.76

some intel coned parts Intel

I also ordered a MH8228, Intel 8228 Clone, SYSTEM CONTROLLER at £0.47p
GeraldSommariva is offline  
Old 16th May 2021, 8:16 am   #174
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by ortek_service View Post
Quote:
Originally Posted by GeraldSommariva View Post
The other problem I am having is with the keyboard putting -12v on the data lines, also the strobe out from the keyboard is sitting at a strange 2.1v.


The keyboard is from Micro Switch and the encoder chip is I think SW20457k
That does sound rather serious - Having -12V on the datalines is unlikely to be good for anything else connected to these, so could cause lots of damage.
And is most likely due to a short to -12V rail in an IC that uses this power rail, on the keyboard.
Although it could be due to the loss of a positive supply rail, if the IC's using these are biasing the lines relative to that

I can't seem to find any info on that obscure Keyboard controller / encoder IC. I wonder if it's equivalent to a more stand AY-x-xxxx one or whether it's a custom microcontroller etc.
And the keyboard itself is now rather uncommon - do you have any documentation that came with it? - All I can find is:
https://deskthority.net/wiki/Micro_Switch_SD_Series
Back on the Keyboard I did some more testing some of the Data lines from the keyboard encoder go from +5v to 0v and some go from +5v to -12v as you press the keys. i made a drawing of the keybaord.

sd-keyboard-circuit-diagram


Would it be usual to have pull down resistors from the encoder output to -12v ?

I moved the resistors so they pull down to 0v, now all outputs go from +5v to 0v as I would expect. This is strange as the keyboard was working with the resistors pulling down to -12v.
GeraldSommariva is offline  
Old 16th May 2021, 9:43 am   #175
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: The Transam Triton Personal Computer

What is the input circuit on the Triton? If its standard ttl or even ls ttl then 2.7 k pull down might not be strong enough to pull down to less than 0.8 v. Ideally you want them to pull down to 0.4v or less. Most of the ttl inputs had diodes to protect the input from undervoltage, so the 2.7k pull down to -12v is probably going to be ok as the 2.7 k resistor would limit the current into the input diodes of the ttl.
Mark1960 is online now  
Old 16th May 2021, 12:25 pm   #176
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by Mark1960 View Post
What is the input circuit on the Triton? If its standard ttl or even ls ttl then 2.7 k pull down might not be strong enough to pull down to less than 0.8 v. Ideally you want them to pull down to 0.4v or less. Most of the ttl inputs had diodes to protect the input from undervoltage, so the 2.7k pull down to -12v is probably going to be ok as the 2.7 k resistor would limit the current into the input diodes of the ttl.
The input to the Triton is via IC49 a 74LS244 Triton ports circuit diagram


I found a problem with this chip on my Triton pin 11 (strobe) was bing passed through to pin 9 (DB8) even when both select pins are high. This had the effect of holding DB8 near zero when the Keyboard was plugged in, that took me some time to find when checking out the data bus.
GeraldSommariva is offline  
Old 16th May 2021, 1:10 pm   #177
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by GeraldSommariva View Post
>>
>>
Back on the Keyboard I did some more testing some of the Data lines from the keyboard encoder go from +5v to 0v and some go from +5v to -12v as you press the keys. i made a drawing of the keybaord.

sd-keyboard-circuit-diagram


Would it be usual to have pull down resistors from the encoder output to -12v ?

I moved the resistors so they pull down to 0v, now all outputs go from +5v to 0v as I would expect. This is strange as the keyboard was working with the resistors pulling down to -12v.

What is even stranger is that any of the outputs only went down to 0V, if they all had pull-downs to -12V - I'd expect all to go down to -12V

I presume it was done that way as this keyboard wasn't actually designed for the Triton etc, and so maybe it did actually work OK for a while like this, as resistors would have limited the current.

If all outputs do go down close to 0V OK, with pull-downs to 0V, then maybe this controller IC is fairly-OK after all, which would be rather handy.
Although maybe some of it's output-circuitry is quite fully-working, but it maybe possible to work-around that
- If you feed into a 74HC744 buffer etc. 74HC-series logic-gate, then that will have < 1/3 x 5V = 1.7V Low-threshold and > 2/3 x 5V = 3.3V High-threshold, which will probably be OK, if the keyboard outputs really go all the way to +5V (TTL often only goes upto 3.5V, hence standard 74LS etc. usually has thresholds shifted down a bit and don't have as good noise-immunity of CMOS levels.
ortek_service is offline  
Old 16th May 2021, 1:48 pm   #178
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Is the Encoder IC and PMOS device its supplied with +5v and -12V ?

I found this on PMOS Logic From Wikipedia:-

Most PMOS integrated circuits require a power supply of 17-24 volt DC.[20] The Intel 4004 PMOS microprocessor, however, uses PMOS logic with polysilicon rather than metal gates allowing a smaller voltage differential. For compatibility with TTL signals, the 4004 uses positive supply voltage VSS=+5V and negative supply voltage VDD = -10V.[21]
GeraldSommariva is offline  
Old 16th May 2021, 2:36 pm   #179
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: The Transam Triton Personal Computer

Yes, it does appear that it's a PMOS device (Some of the General Instruments ones also needed voltages other than just single +5V).
And +5 - (-12V) = 17V, so would fit in with that voltage range (rather than slightly-narrower Intel total 15V difference).
PMOS is inherently 'Positive-Ground', so often has Vss (Usually 0V/Ground on later NMOS & CMOS) connected to highest +5V voltage, and other supply pins are negative relative to that. But the datasheets can sometimes be rather unclear over this.

This encoder IC does seem to also have a 0V 'reference' pin as well - Whereas some like the original NS SC/MP-I CPU, and early EPROM's didn't actually have a 0V pin (So must have internally-biased their outputs at levels to give the standard 0V/ 5V 'TTL' levels that most on-board data-buses ran at).

Some slightly-later EPROM and DRAM IC's, were 3-rail devices, so as well as using +5V & 0V, these also needed 2 extra supplies - Often instead using +12V and -5V to get 17V total.

Last edited by ortek_service; 16th May 2021 at 2:44 pm.
ortek_service is offline  
Old 16th May 2021, 4:41 pm   #180
GeraldSommariva
Pentode
 
GeraldSommariva's Avatar
 
Join Date: Mar 2021
Location: Liphook, Hampshire, UK
Posts: 125
Default Re: The Transam Triton Personal Computer

Quote:
Originally Posted by ortek_service View Post

If all outputs do go down close to 0V OK, with pull-downs to 0V, then maybe this controller IC is fairly-OK after all, which would be rather handy.
Although maybe some of it's output-circuitry is quite fully-working, but it maybe possible to work-around that .
Having done more testing I have produced the following Output Table and its Looks like the controller is working as you predicted.

The ASCII values look correct and the shift and CTL function seam to work, I been feeding the outputs from the encoder in to a 74LS244 and reading them with by PIC processor Raspberry PI combo.
GeraldSommariva is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:58 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.