Restoring the BT Cheetah 85

Anything that isn't support related

Moderator: M1ECY

Antworten
Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Restoring the BT Cheetah 85

#1

Beitrag von M1ECY » So 3. Nov 2019, 15:57

Some of you are probably sick of hearing about the mad happenings in my radio shack, but for those of you that are slightly interested, please see some of the ramblings from me regarding the ongoing process of restoring the BT Cheetah 85 Teleprinter.

I must firstly thank Rainer, Henning, and Werner for their help so far-each of you have played a part in getting this far.

So, to set the scene.

The British Telecom Cheetah 85 and 87 Teleprinters are both very similar electrically, they were a joint venture with SEL Lorenz, and the Lo3000 is almost exactly the same machine in a slightly different case. This machine appeared on eBay earlier this year, and was a fairly good price (£75GBP) It was located in the North West of England, so required a 3 hour drive to collect it - At this point it is fair to say that my long suffering Wife accompanied me for this trip, and she does not share my enthusiasm for all things old and electrical.

My original plan was to use this machine as a source of parts to repair my other Cheetah - Henning and Werner had been kind enough to supply me with a spare set of Lo3000 boards to try to get this machine working, but this was not proving to be a good solution. I decided that before I disturbed anything, I would try a short power up test, to see if there was any life at all - the machine had sat in my workshop for about 5 months, so should have dried out from it's previous agricultural storage. Power was applied, and one finger was left on the mains switch just in case the power supply decided to turn into a smoke grenade.

Much to my disappointment the power supply behaved correctly, and the Cheetah started making encouraging bleeps, indicating that it would like a fresh roll of paper - for some reason the reset button would not shut the beeping up, so I had a look at the screen.

It reported, quite correctly that the battery was faulty, and all options had been lost.

So, it seems much fairer at this point to allow this new machine to live.

Happy that this proved the machine was at least partly viable, I then shut everything down and removed the as yet un-detonated smoke grenade - first order of business would be to prevent any magic smoke from escaping, as Lo3000 owners will know, the power supply module is not the most reliable thing after some 15+ years of service, and then a 25 year retirement period.

Here in the UK we are well versed in the activities of the RIFA class X and Class Y supression capacitors, and have considerable experience of their behaviour when woken from an extended sleeping period. As much as I enjoy the excitement of a smoke filled room, these components were the first thing to be removed from the PSU.

Based on the experience of Patrick, and my previous knowledge of Switch Mode Power Supplies, I have also ordered a full set of replacement electrolytic capacitors - Most of the components in this PSU are still available, but with the advance in technology, they are all a lot smaller! Wound components are all obsolete, so prevention is a major part of this area of restoration - Where safety or reliability are a concern, originality is sacrificed, and modern components are used.
IMG_20191027_161257_463 (1).jpg
The PSU is different from the Lo3000 in that all the components are on the one PCB - this module appears to be a replacement that made it's way into the machine right at the end of the lifespan of Telex in the UK
Du hast keine ausreichende Berechtigung, um die Dateianhänge dieses Beitrags anzusehen.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Re: Restoring the BT Cheetah 85

#2

Beitrag von M1ECY » So 3. Nov 2019, 16:50

Whilst waiting for all the new parts for the power supply to arrive I continued work on the machine, using the spare, and already repaired PSU from my other Cheetah.

This allowed me to find some extra problems

1, The print carriage is jammed.
2, The battery has leaked quite badly.
3, I cannot access the options menu or list functions.

The carriage was quite easy to solve - there was a lot of dust and dirt on the guide rail, this had allowed some slight corrosion of the guide to jam up the printhead carriage. After cleaning the dirt and corrosion off, I lubricated the bar and guide bearings with a couple of drops of hydraulic oil, and re tested the action of the print head - now with each power up the head moved and reset position as expected.

After several attempts at getting access to the operator and engineer menu, I went to the forum resident Lo3000 expert (Rainer) with some questions - He kindly removed his battery from his machine, and showed me the power up sequence, and results of the key press sequences - all of which should be the same for the Cheetah.

There is a difference between the Lo3000 and Cheetah for access to the "Engineer" mode - there is no "Engineer Plug", there is however a switch on the CPU board that performs this function, however, no matter what I tried, I could not make anything happen.

At this point, I started to wonder if there was a problem with the keyboard - everything else seemed to be working correctly, so I really had nowhere else to go - The poor old Cheetah 87 was quickly raided and it's keyboard removed. Plugging this into the patient gave some promising results - I now was able to enter the elusive Engineer mode, and operator function list.

Initially I wondered if there was a problem with either the eprom, or keyboard driver chip, so I swapped these over in turn, and in all cases the replacement keyboard worked - this meant that both of these parts were all working from the original keyboard.

I then tried dismantling the keyboard, and cleaning the contact faces - no improvement was gained - time for an internet search.....

The keys on the Cheetah are manufactured by Keytronic - they are also used on vintage computers like the Apple Lisa and I suspect things like the BBC micro too. There are countless threads and discussions on the internet that mention the restoration of these keyboards.

So I am dealing with a capacitive keyboard fault - behind the keys there is a composite pad, made up of a thick mylar disk, some open cell foam, and another disc of foil backed mylar film - this second disc is pressed onto the circuit board, and makes a capacitive path between the two traces on the board.

In the case of the faulty keyboard, all of the metallising of the mylar film has failed, and the open cell foam has decomposed - this means that there is no possibility of the keys working. Again Google is your friend, and a company in the USA manufactures replacement composite pads - www.texelec.com supply bags of 105 pads - these have been ordered, and I will report back on their performance when they arrive.

In the meantime, there is 35+ years of accumulated nastiness in the keyboard, so, what to do - Clean up time!
IMG-20191028-WA0013.jpg
I will not say how long it took to remove that lot, nor will I repeat some of the language I used when they wouldn't come out :)

Being quite fragile, and worried that I may destroy the print on the keys, I did not want to use any solvent, or strong detergent on the keys - Again Google to the rescue - I found a suggestion to use denture cleaning tablets, so a trip to the local supermarket, a bite to eat, and a tube of said denture cleaning product.

All the keys were dropped in a plastic bowl along with 6 tablets, the bowl was filled with cold water, and left for 12 hours to soak.

Once removed and rinsed, the keys were all dried off with a towel, and reinserted into the frame - Top Tip - do not start this task at Midnight!
20191029_213248 (1).jpg
20191029_213301 (1).jpg
20191029_223253 (1).jpg
At around 01:30 I had managed to get all the keys back in the right place, and I passed the keyboard to the foreman for inspection.
20191029_223313 (1).jpg
Du hast keine ausreichende Berechtigung, um die Dateianhänge dieses Beitrags anzusehen.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Re: Restoring the BT Cheetah 85

#3

Beitrag von M1ECY » So 3. Nov 2019, 17:28

While awaiting the arrival of the ever increasing parts list, I took a quick look at the original battery module.

It is in a very poor state - every soldered joint is corroded, and there is corrosion all over the PCB - I tried replacing the batteries with a more modern unit, and reflowed all the joints, but the unit is still faulty - I will try to repair it at a later date.

I know the Central Teleprinter Stores in Blumenhagen has these items in stock, I already have one of these new old stock items for the other Cheetah, so for now I am using this in the machine.

Plugging this battery in to the machine has allowed me to program some options, and for the memory to retain them. Certainly a positive start.

At this point I had a machine that was around 50% functional.

Onto getting the printer to work.

I could send items to the printer, and it made all the right noises, and fed paper through without a problem, but it was missing one important thing - No Ink!

For reasons that i do not fully understand, ribbons for this machine are no longer available in the UK, after making enquiries, it appears that there is a very limited stock of ribbons at the Central Teleprinter Stores, and there is also a small stock at the Münster-Sarmsheim outpost of the Teleprinter Support Network, who also advised that there was a re-inking service available in Germany - Not wanting to use up these valuable stores (and being an impatient creature) I thought I would have a go at re-inking the ribbon myself.

A long time friend of mine works in a stationery shop in town, so I visited her to purchase a bottle of "stamping ink" - a 50ml bottle cost £1.50, I have to be honest, I was not expecting any chance of this working, but had nothing to loose.

Upon returning home, I made a start. Splitting the case for the ribbon is a simple task, just a small screwdriver is needed, and it separates quite easily. Not having done this before, I added a few drops of ink....

It didn't exactly soak in as planned, so I reckoned that i needed to move the ribbon around. Another top tip, :mentor: do not tip the ribbon cassette up without first putting the lid on. It all falls out, lands on the floor in a tangled inky mess, and takes ages to put back in :headshot:

Once I had managed to get the ribbon back in the cassette, and cleaned up the workshop, me and all the tools and implements used in the great ribbon race, I continued refill the ink - one benefit of the ribbon falling out was that everything was nice and loose now, and the ink soaked in quite well.

I got bored very quickly of using the manual winder to move the ribbon around - the drive is a small cross in the rear of the cassette - strangely a No 2 Pozidrive screwdriver fits well. Aha, an idea. A cordless screwdriver really speeds up the moving of the ribbon - you can easily see the sections of ribbon that are wet with ink, and the dry patches can be attended to with the bottle of ink. about 30 minutes of winding and dripping made for a completely reinked ribbon.

Once installed in the machine (and a good clean up of the war zone :) )testing could commence.
IMG-20191029-WA0000.jpg
Du hast keine ausreichende Berechtigung, um die Dateianhänge dieses Beitrags anzusehen.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Re: Restoring the BT Cheetah 85

#4

Beitrag von M1ECY » So 3. Nov 2019, 17:58

Up till this point, I had given little consideration to connecting the machine to I-Telex, The Cheetahs I have here, both have an external box called "Unit Telegraph No 55a" This is a clever little box that converts a V.21 signal into the double current signalling that the Cheetah needs.

Unfortunately, I-Telex does not (Yet) support V.21, nor does it (Yet) support double current systems, so connecting this to the network was going to be difficult.

Alf has managed to make a converter for the Creed 444 at The Signals Museum, Henlow, and it works very well, in time I will build one of these to allow for double current operation. Henning is also working on a modified analogue board for I-Telex that will allow V.21 operation. none of these solutions are to hand at present.

Remember the Lo3000 boards I mentioned earlier? Well in amongst these boards was a TW39 LAT card (the LAT card is what allows the teleprinter to connect to the outside world) With nothing to loose, I swapped out the original double current board for the new card, and disconnected the UT55a. Pins 1 and 4 carry the two wire TW39 signal on the filter board, so I made up a new cable to connect to I-Telex.

As ususal if there are two possible polarity solutions, I will pick the wrong one first, so on swapping the A and B connections around, I tested the connection.
IMG_20191102_125819_015 (1).jpg
Aha, my 50% functional machine quickly became a 75% functional machine - Having no actual experience of a working Cheetah, and to my knowledge, there being no other working example in the UK I was quite happy. Onto some on line testing.
VID-20191102-WA0001.mp4
Du hast keine ausreichende Berechtigung, um die Dateianhänge dieses Beitrags anzusehen.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Re: Restoring the BT Cheetah 85

#5

Beitrag von M1ECY » So 3. Nov 2019, 18:43

So, What looks like a fully functional machine, a few telex exchanges with Alf, Henning, and Johannes, as well as my good friend the DWD Wetter Service, as well as a short test of the new conference room, and all looked good.

Then, the screen failed.

Initial thoughts were not printable here and :klopf: :klopf: :klopf: :klopf: :klopf: was my plan.

After having 5 minutes to collect my thoughts, I swapped the monitor for the one from my other machine - all still working, which is good news.

So, I pull the original monitor apart - it is very dirty inside, and because the unit works upside down, all the heat buildup is collected by the PCB - this is a poor design, but it must work OK, after all, up till half an hour ago it still works :)

Initial concerns are for the flyback transformer, the one item that would effectively kill the monitor off.

A night to sleep on it, and some testing today
20191103_123450x (1).jpg
The Flyback transformer passes a ringing test, so it is onto the driver circuitry - The H switching transistor tests OK, so it must be before that. Really difficult fault finding without a circuit diagram - Perhaps someone here has one?

More to do......
Du hast keine ausreichende Berechtigung, um die Dateianhänge dieses Beitrags anzusehen.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Re: Restoring the BT Cheetah 85

#6

Beitrag von M1ECY » So 3. Nov 2019, 20:58

Spent another hour looking at the monitor fault - there is no drive for the line output stage - not having a circuit diagram is really hurting now - but this is good news really, hopefully the fault will not be anything too difficult to repair.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Benutzeravatar

Topic author
M1ECY
Rank 8
Rank 8
Beiträge: 161
Registriert: Fr 11. Aug 2017, 13:13
Wohnort: Hitchin Hertfordshire, UK
Hauptanschluß: 459724

Re: Restoring the BT Cheetah 85

#7

Beitrag von M1ECY » Do 21. Nov 2019, 21:50

Now that the machine is basically running, a little update is in order.

Monitor:

For now I have swapped the monitor from the Cheetah 87 into this machine - faultfinding on the monitor is going nowhere fast, and really needs a schematic.

Power Supply:

I had already almost completed a refurbishment of the power supply for the Cheetah 87, just needed the two primary side electrolytic capacitors to be changed. Thanks to the information that Patric shared earlier this year, I was able to buy a small stock of these components from Germany. This evening I continued to fit these last two components. Because it had been some time since I last worked on this supply, I had completely forgot the layout - See below.
20191121_183155 (1).jpg
Seeing as these machines were developed with SEL Lorenz, and were marketed as the ITT3000, it is surprising to see the differences between this and the Lo3000 power supply. Even more interesting is the difference between the original supply for the Cheetah 85!
20191121_183205 (1).jpg
I have a lot more work to do to the original 85 power supply, and I am awaiting a delivery of British capacitors from Richelt :llach: So, this evening I completed the repairs to the supply from the 87.
20191121_185514 (1).jpg
20191121_185523 (1).jpg
I need to invest in a single phase power monitor to see if the power consumption of the machine is close to the manufacturers data, there really isn't much left that can be replaced on an age basis.

There is one annoying fault, that is going to be tricky to solve - after about 20 minutes of inactivity the TP goes into alarm, and displays "Fundamental Fault" There is of course a workaround - I could just use an FSG to kill the power from the machine - I may look into designing a small unit for the more modern machines - because you dial either from memories, or from the keyboard, it is a waste of a full FSG unit.

Until I have full trust in the machine, it is only powered up while I am near it - Spare parts are almost impossible to get for these machines in the UK.

More to do, but I look forward to exchanging a few more messages on the Cheetah soon
Du hast keine ausreichende Berechtigung, um die Dateianhänge dieses Beitrags anzusehen.
459724 SEAN G - Te Ka De FS200 - Working 24/7
666507 JONAB G - BT Cheetah 85 - Working sometimes
350470 Ersatz G - J-Term - Working 24/7
BT Cheetah 87 "Kranken" a spares donor now :(
Siemens T68d - A work in progress
BT Super Cheetah - Functional, but awaiting SVCF I-Telex Card

Antworten

Zurück zu „General Discussion“