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 21st Jan 2023, 5:42 pm   #1
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default PET 3016 Datasette problems

Due to an entirely self-inflicted and somewhat embarrassing release of the magic smoke, I have a problem with my datasettes on my PET.

I soldered up a Stupid PET Tricks device, and got cocky and the smoke was let out of that, due to a short on my soldering. The device plugs into datasette port 1 for power and the User Port for the device itself.

It seems to have had the side effect of making datasette port 1 unreliable and port 2 not work at all.

So the symptoms I now have are a known working cassette and drive will sometimes work on port 1 but when I plug it into port 2, it is recognised (becuase when I press play, the PET says SEARCHING) but there is no power to the spindles so the tape does not move. Even when not using a LOAD command, none of the REW/FF/PLAY buttons do anything on the device itself.

I have swapped the 6520s, swapped the 6522 from my disk drive and tried a Tynemouth board to effect a swap of the 6502 and the ROMs and RAMs - all to no avail.

So I guess I've done something bad to the motherboard itself.

I've taken measurements of datasette 2 pins when trying a LOAD"",2 and I have low voltage (0.3v) on pin 4 (MOTOR) so I'm guessing that's where to start. I also get a lower than I expect voltage on Pin 4 of datasette 1 (4.4V) which I am guessing is connected to the intermittent loading on that device.

I get 5V on ports 1 and 2 pin 2 so that's OK.

Looking at the schematics, I'm guessing that there's something wrong with the transistors at Q1 and/or Q4 - but I have no idea how to test them.

Help please?

Thanks as ever.

Colin.

Last edited by ScottishColin; 21st Jan 2023 at 6:11 pm.
ScottishColin is offline  
Old 21st Jan 2023, 6:55 pm   #2
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: PET 3016 Datasette problems

I've not got the schematic to hand / can't remember which version you had, but testing the transistor can probably be done without that for now, in-circuit with a DMM set to 'Diode-Test'
Just need to look-up the transistor numbers to find pinout and whether they are NPN or PNP.
It should just be a case of (with PET powered-off) putting one probe (Red for NPN, Black for PNP for a DMM) on the base and the other probe on emitter and then collector. And you should get a reading of around 0.6V to 0.8V, giving you a quick-check to see if any junction is open or short.

But to do a more extensive test, then probably easiest to remove transistor and connect it to transistor test socket on DMM (assuming yours has this), to measure the hFE (DC-gain) of it. It may be possible to do it in-circuit, with some extension leads from DMM sock, but sometimes need to desolder a lead or two to isolate the rest of the circuitry from affecting measurement.

Alternatively, you could just measure DC-voltages of both transistors, with Motor signal enabled & disabled, to check they are switching OK / find where the fault lies.
ortek_service is offline  
Old 22nd Jan 2023, 2:24 pm   #3
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: PET 3016 Datasette problems

Taking Port2 which doesn't work at all first (because definite faults are usually easier to fix than half-faults), what voltage do you see across diode VR1 when the cassette motor should be running?

With a known working datasette connected to port 2 and the PET and the datasette both in a state in which the motor should be running, what voltages (with respect to 0V) do you see on the collector, base and emitter respectively of TIP29 transistor Q1? The attached TIP29 pinout diagram (borrowed from an online source) should assist with pin identification.
Attached Thumbnails
Click image for larger version

Name:	TIP29-pinout.jpg
Views:	25
Size:	73.2 KB
ID:	271991  
SiriusHardware is offline  
Old 22nd Jan 2023, 4:04 pm   #4
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: PET 3016 Datasette problems

Incidentally, your friends at Cricklewood appear to have anticipated your future need to fix this problem and have both the TIP29 (Q1, Q4) and the 2N4401 (Q2, Q5) at reasonable prices:-

TIP29:

https://www.cricklewoodelectronics.com/TIP29.html

2N4401:

https://www.cricklewoodelectronics.com/2N4401.html
SiriusHardware is offline  
Old 22nd Jan 2023, 5:02 pm   #5
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: PET 3016 Datasette problems

VR1 with a LOAD "",2 and PLAY pressed on the datasette (which slowly moved this time for some reason - it hadn't moved at all before):

7.9V

Q1 is handily marked B C E on the motherboard:

B 3.2V
C 9.6V
E 2.6V

The TIP29 on Q1 and Q4 are riveted to the motherboard - that's going to be fun replacing them. They also look like they have some kind of white material between them and the motherboard; would that be some kind of thermal paste? Or is that an indication they've broken?

Colin.


Quote:
Originally Posted by SiriusHardware View Post
Taking Port2 which doesn't work at all first (because definite faults are usually easier to fix than half-faults), what voltage do you see across diode VR1 when the cassette motor should be running?

With a known working datasette connected to port 2 and the PET and the datasette both in a state in which the motor should be running, what voltages (with respect to 0V) do you see on the collector, base and emitter respectively of TIP29 transistor Q1? The attached TIP29 pinout diagram (borrowed from an online source) should assist with pin identification.
ScottishColin is offline  
Old 22nd Jan 2023, 5:29 pm   #6
Realtime
Hexode
 
Join Date: Jan 2021
Location: Ashford, Kent, UK
Posts: 318
Default Re: PET 3016 Datasette problems

That's almost certainly thermal paste. If they're riveted down then they must need a good thermal path to (I would assume) the PCB copper planes. Best to clean off the old stuff and apply new. Cricklewood also have thermal paste

https://www.cricklewoodelectronics.c...a-Syringe.html

Enough to last a lifetime for most people.
Realtime is offline  
Old 22nd Jan 2023, 6:07 pm   #7
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: PET 3016 Datasette problems

I'm hoping that the b-c-e connections for Q2 (2N4401) are also helpfully marked.

If they are could you please report the voltages on Q2 collector, base and emitter under the same test conditions you used when measuring the Q1 voltages.

At the moment it is by no means certain that Q1 is faulty as the base-emitter voltage difference looks about OK, the problem being that the voltage going to the base should be quite a bit higher, hence the focus now on Q2.

Last edited by SiriusHardware; 22nd Jan 2023 at 6:15 pm.
SiriusHardware is offline  
Old 22nd Jan 2023, 8:44 pm   #8
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: PET 3016 Datasette problems

Q2:

B 8.2V
C 10.2V
E 3.6V

Colin.
ScottishColin is offline  
Old 22nd Jan 2023, 9:00 pm   #9
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: PET 3016 Datasette problems

Looks like Q2 is faulty. Voltage between base and collector is far too high.

I think it might be possible to substitute any general purpose npn transistor if you have any rather than ordering one.

Last edited by Mark1960; 22nd Jan 2023 at 9:01 pm. Reason: Correct spelling
Mark1960 is online now  
Old 22nd Jan 2023, 9:03 pm   #10
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: PET 3016 Datasette problems

For a healthy Silicon NPN transistor it should be 'impossible' to have such a large positive voltage difference across the b-e junction - should be no more than about 0.6V, so I am going to say that Q2 is faulty. Do you have any transistors of any sort in your parts bin?

Edit: Crossed with Mark.
SiriusHardware is offline  
Old 22nd Jan 2023, 11:19 pm   #11
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: PET 3016 Datasette problems

We have a choice - see attached photo. Are any better than others for this problem?

Colin.
Attached Thumbnails
Click image for larger version

Name:	PXL_20230122_221605052.jpg
Views:	33
Size:	73.4 KB
ID:	272021  
ScottishColin is offline  
Old 23rd Jan 2023, 12:26 am   #12
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: PET 3016 Datasette problems

Probably just use the BC548, but check manufacturers data sheets to make sure the pinout matches what you are replacing.

BC327 is pnp
BC337 is npn
BC517 is a darlington pair
BC547 is higher voltage, but not needed
BC549 is low noise
BC550 is higher voltage and low noise
BC556 to BC558 are pnp.
Mark1960 is online now  
Old 23rd Jan 2023, 12:35 am   #13
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: PET 3016 Datasette problems

The original 2N4401 is an NPN rated at 40V 0.6A 0.625W. See: https://www.onsemi.com/pdf/datasheet/2n4401-d.pdf

The BC337 is NPN, rated at 45V 0.8A 0.625W. See: https://www.onsemi.com/pdf/datasheet/bc337-d.pdf
is probably the best choice out of these, as most of the other NPN ones are only rated at 100mA collector current
- Just in case that is important and they are running collector current higher than that.

The BC517 is NPN, rated at 30V 1.2A 0.625W could be another choice, but it is a darlington so will saturate at a rather higher collector-emitter voltage if that might be a problem. It will also have a much higher gain, but that probably won't be an issue.
https://www.mouser.com/datasheet/2/149/BC517-888613.pdf

Note: The original 2N4401 has the pinout reversed relative to the BC337 & BC517.
- So if using either of those (or the BC54x ones: https://www.sparkfun.com/datasheets/...ents/BC546.pdf ) , then will need to fit the opposite way round to the originally-used 2N4401 that has a less-common pinout arrangement.

BTW, You do occasionally find TO-220 transistor riveted in place, so have to gently (slowly) drill the head off the rivet with a large drill bit. Yoy may get away with reusing some of the old thermal-grease, as only need a thin smear, nbuy any PC processor heatsink compound should also be OK if you have any. If there's any electrically-insulating mica etc. washer, then have to be careful to not damage it / refit it (or replace with less-fragile silicone rubber type), as well as any insulating bush through the tab - As well as find a suitable screw, washers & nut to replace the rivet - But hopefully these have all survived OK and so don't need to change.

Last edited by ortek_service; 23rd Jan 2023 at 12:55 am.
ortek_service is offline  
Old 23rd Jan 2023, 12:56 am   #14
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: PET 3016 Datasette problems

I would say certainly not the BC517 Darlington, for the reason that the 2N4401 and TIP29 already are a Darlington pair. This is the relevant circuit section:-

http://www.zimmers.net/anonftp/pub/c...N/320349-3.gif

We are currently looking at Q1 / Q2 which control the motor drive to cassette port 2. Given the potential for mix-ups over the different pinouts I am not sure if Colin shouldn't just order exact replacements, but if he is willing to have a go bearing in mind the difference in pinouts then the BC337 sounds like a good choice for the reasons Owen suggested - higher current rating is probably more important than higher voltage rating or gain, given that a Darlington pair has inherently high gain.

As Owen said, we don't necessarily think that the TIP29 in this channel has failed so leave well alone for now, don't be trying to drill out the rivets as there is obviously risk associated with that, whether a bit-slip or shards of fine metal swarf disappearing off and getting stuck under IC sockets to cause inexplicable faults.
SiriusHardware is offline  
Old 23rd Jan 2023, 1:17 am   #15
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: PET 3016 Datasette problems

Yes, looking at the circuit and that they clamp the 2N4401 base at 7.5V with the VR1 zener, then using a darlington would mean lower output voltage and more voltage / power across the TIP29 output transistor.

The TIP29 only has an hFE of 15 min at full 1A load, so needs >67mA base-drive from Q2/Q5 to ensure it is saturated (although hopefully cassette motor doesn't draw that much). So the 100mA rated BC54x ones would normally be just about OK.
But if motor gets stalled / there is an accidental short on Motor voltage output, then hopefully the BC337 is much more likely to survive (and may also be less likely to cause TIP29 to fail, if providing more base drive to it and not cause that to overheat due to excess voltage-drop if not saturating)

Swapping the 2N4401 to a BC337 should be fairly-simple, as just need to fit it rotated 180degrees as both at least have the base in the centre (with no complicated re-bending of the leads that you have with 'L' suffix / other types with collector in the centre)
ortek_service is offline  
Old 23rd Jan 2023, 4:18 pm   #16
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: PET 3016 Datasette problems

Thanks all.

Problem resolved by replacing Q2 with a BC337.

Just having in internal debate now about whether I should buy a 2N4401 so I can put it back to 'original' parts.

Then onto trying to fix the Stupid PET tricks device to get it working without the further release of the magic smoke.

Colin.
ScottishColin is offline  
Old 23rd Jan 2023, 7:26 pm   #17
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: PET 3016 Datasette problems

If you want you can repeat the same process for Datasette port 1,

With the datasette connected to port 1 and the PET and datasette in a state where tape 1 should be rolling,

Measure the voltages:-

-Across diode VR2.
-On collector, base and emitter of Q5 (2N4401).
-On collector, base and emitter of Q4 (TIP29).

The likelihood is that this channel is damaged in the same way as the other one, but just not quite as badly. If you measure the '#2 Cass Motor' voltage on Port2 when the tape is running it will probably be no less than 6V.

If the voltage going out of the '#1 Cass motor' terminal to the datasette is any less, that motor drive output is probably still half damaged and still needs to be fully repaired.
SiriusHardware is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:12 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.