Atari 1050 Track 0 sensor repair

[TGB1718]

This may have been covered before, but I couldn't find anything here.

Basically I have 2 1050 drives, one US doubler and the other a Happy (not so happy).

The Happy one made the usual noises at power up, but would not boot/read disks, I removed the cover and found the

head was only moving forward during power up until it hit the end of it's travel.

 

I did all the usual stuff, checking PSU voltages, noise etc. with a scope (I had replaced all the Cap's on these drives a few years ago)

I checked the signals to the stepper motor and although they didn't quite match the waveforms in the service manual, they were close

enough for me to believe the signals were good, I also checked the motor with a meter and again it seemed OK.

 

I then checked the track 0 sensor by pushing the head in and out of the sensor, no signal change was found.

I 'Googled' Photo Interrupter and found several on Ebay, I check the data sheets and this one seemed perfect for the job.

 

https://www.ebay.co.uk/itm/Transmissive-Photo-Interrupter-Phototransistor-Panel-Chassis-3-18-mm-1-27/351224553074?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

 

I know its an ebay link, so here's the datasheet link too

 

https://uk.farnell.com/tt-electronics-optek-technology/opb840w51z/opto-switch-slotted/dp/1497932?gclid=CjwKCAiAqOriBRAfEiwAEb9oXW_9wBvAmbAS-ZkpLt9X6O3nC9iL4lKN5xcbMTqsqlFPEfRU_78PCBoCa-IQAvD_BwE&mckv=sKzEcuqhA_dc|pcrid|202041827980|kword|opb840w51z|match|p|plid||slid||product||pgrid|5276988248|ptaid|kwd-10802725947|&CMP=KNC-GUK-GEN-SKU-MDC

 

It was a perfect fit, I cut the wires about 1/2 way to the connector and cut the wires on the new sensor to fit and soldered them (using heatshrink sleeving over the new joints.

 

Important;!!! the wires to the internal LED are reversed i.e. 1 and 2 on original sensor are connected to 2 and 1 on the new sensor.

Also it's worth marking the drive metal where the sensor is currently located (just to give a rough idea of the starting point.

 

Powered up the drive and success, the head moved back to the track 0 position.

 

Next powered up my 130XE with a diagnostics disk via Respqt on a Raspberry Pi as disk 1

Once booted I removed the virtual drive and connected the Happy Drive with it selected as Drive 1

I ran the Track 0 alignment test with a known working Diagnostics disk from my US Doubler drive.

 

Tighten the screw on the sensor to the point where using a small screwdriver so you can just move the sensor with the aid of a screwdriver, luckily there are suitable

high points on the drive casing to lever the screwdriver against.

After a few iterations of 'BACKWARD'/'FORWARD' messages while moving the Track 0 Sensor I found the 'Sweet Spot' by very small

movements of the sensor. (I found it odd that no message saying 'OK' came up.

 

I fully tightened the sensor, left the Diagnostics disk in, re-booted the 130XE and to my relief it booted fine.

 

I am posting this mainly to show that there is a replacement part that's a perfect fit.

Hope this helps someone.

Have also attached the datasheet

[TGB1718]

File did not attach in original post, here it is

2331364.pdf

[E474]

Thanks very much: how-to article + where to get replacement part(s) + data sheet is really helpful.

[+Nezgar]

I haven't yet had a drive that needed this repair but I very much appreciate your findings and posting it here. Knowing it's here for reference is reassuring, and will be useful for others as original parts become harder to find.

[tep392]

I'm glad I found this.  Tonight I learned that accidentally swapping the J1 and J10 connectors will kill the track 0 sensor.

 

[tep392]

Turns out the World Storage mechs use a different package for the sensor, so the OPB840 type wont fit.  I discovered that it can be pulled apart, so I'm going to try replacing the bad LED and see if that fixes it.  I was able to check the photo transistor works by shining my phone's light into it.

[+kheller2]

WST mechs use lots of different things from Tandon mechs.  

[macsonny]

On 2/9/2019 at 12:16 AM, TGB1718 said:

This may have been covered before, but I couldn't find anything here.

Basically I have 2 1050 drives, one US doubler and the other a Happy (not so happy).

The Happy one made the usual noises at power up, but would not boot/read disks, I removed the cover and found the

head was only moving forward during power up until it hit the end of it's travel.

 

I did all the usual stuff, checking PSU voltages, noise etc. with a scope (I had replaced all the Cap's on these drives a few years ago)

I checked the signals to the stepper motor and although they didn't quite match the waveforms in the service manual, they were close

enough for me to believe the signals were good, I also checked the motor with a meter and again it seemed OK.

 

I then checked the track 0 sensor by pushing the head in and out of the sensor, no signal change was found.

I 'Googled' Photo Interrupter and found several on Ebay, I check the data sheets and this one seemed perfect for the job.

 

https://www.ebay.co.uk/itm/Transmissive-Photo-Interrupter-Phototransistor-Panel-Chassis-3-18-mm-1-27/351224553074?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

 

I know its an ebay link, so here's the datasheet link too

 

https://uk.farnell.com/tt-electronics-optek-technology/opb840w51z/opto-switch-slotted/dp/1497932?gclid=CjwKCAiAqOriBRAfEiwAEb9oXW_9wBvAmbAS-ZkpLt9X6O3nC9iL4lKN5xcbMTqsqlFPEfRU_78PCBoCa-IQAvD_BwE&mckv=sKzEcuqhA_dc|pcrid|202041827980|kword|opb840w51z|match|p|plid||slid||product||pgrid|5276988248|ptaid|kwd-10802725947|&CMP=KNC-GUK-GEN-SKU-MDC

 

It was a perfect fit, I cut the wires about 1/2 way to the connector and cut the wires on the new sensor to fit and soldered them (using heatshrink sleeving over the new joints.

 

Important;!!! the wires to the internal LED are reversed i.e. 1 and 2 on original sensor are connected to 2 and 1 on the new sensor.

Also it's worth marking the drive metal where the sensor is currently located (just to give a rough idea of the starting point.

 

Powered up the drive and success, the head moved back to the track 0 position.

 

Next powered up my 130XE with a diagnostics disk via Respqt on a Raspberry Pi as disk 1

Once booted I removed the virtual drive and connected the Happy Drive with it selected as Drive 1

I ran the Track 0 alignment test with a known working Diagnostics disk from my US Doubler drive.

 

Tighten the screw on the sensor to the point where using a small screwdriver so you can just move the sensor with the aid of a screwdriver, luckily there are suitable

high points on the drive casing to lever the screwdriver against.

After a few iterations of 'BACKWARD'/'FORWARD' messages while moving the Track 0 Sensor I found the 'Sweet Spot' by very small

movements of the sensor. (I found it odd that no message saying 'OK' came up.

 

I fully tightened the sensor, left the Diagnostics disk in, re-booted the 130XE and to my relief it booted fine.

 

I am posting this mainly to show that there is a replacement part that's a perfect fit.

Hope this helps someone.

Have also attached the datasheet

Thanks for sharing. I had to do something similar to a faulty 1050 drive 12 months ago. Used the exact same sensor but with a bit of modification but worked in the end. Interestingly I purchase 4 sensors (just in case) from MOUSER at the start of the yer. Current delivery estimate is Feb 2023!

 

[SviRu]

On 8/22/2022 at 11:07 PM, macsonny said:

Thanks for sharing. I had to do something similar to a faulty 1050 drive 12 months ago. Used the exact same sensor but with a bit of modification but worked in the end. Interestingly I purchase 4 sensors (just in case) from MOUSER at the start of the yer. Current delivery estimate is Feb 2023!

 

Wasn't plug and play?

[TGB1718]

38 minutes ago, SviRu said:

Wasn't plug and play?

???

[macsonny]

No it wasn’t plug and play as the sensor was not a 1-for-1 fit physically. I had to do a bit of modification to the actual sensor so it would fit in the same place as the previous sensor.

 

Not a difficult job - was just a bit fiddly. I didn’t get the exact same part as Apple used as it was too difficult to source in Australia.