• Please review our updated Terms and Rules here

IBM 5153 (CGA monitor) line voltage range

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
While investigating ways to convert my North American-procured 5153 from "120 VAC" over to 230 VAC, I noticed the design of the internal switching power supply may not actually preclude handling both input voltages. Looking further, I found this quote in the manual on minuszerodegrees.net:

attachment.php


On the back of my particular unit it states 120 VAC and according to the SAM's manual the EMI filtering near the IEC connector has 120 VAC caps but I see that on a pic I took of the PS board a few years ago that it's additional mains filter uses 250 V caps.

Another concern is the degaussing coil and series PTC.

Has anyone operated their 5153 on both mains voltages and what is the marked specification of that unit?

attachment.php
 

Attachments

  • Capture.jpg
    Capture.jpg
    24.8 KB · Views: 1
  • 008.jpg
    008.jpg
    92.8 KB · Views: 1

Malc

Veteran Member
Joined
Jul 16, 2009
Messages
2,223
Location
UK
I thought it went by Model No: ie: 5153001 = 120VAC only, 5153002 / 003 = 230VAC, Ive just had a look at the back of my 5153002 model and the rating is:

50/60 Hz
100 - 250V~@0.95A

I have never tried it on 120V as i don't own a step down transformer.
 

modem7

Veteran Member
Joined
May 29, 2006
Messages
8,474
Location
Melbourne, Australia
Ive just had a look at the back of my 5153002 model and the rating is:

50/60 Hz
100 - 250V~@0.95A

I have never tried it on 120V as i don't own a step down transformer.
My 5153002 is similarly marked.
I just now powered it up using a 240V-to-115V step down transformer, and as expected for the marked voltage range, it worked fine (displayed video as expected).
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
Right ... thanks guys. My unit is only marked "5153."

As you might imagine it takes a great leap of faith to plug a monitor marked "120V~" into 230 VAC but I did that after removing the PS and powering it on the bench against a dummy load. No need to risk the whole monitor.

Much to my relief it worked fine although I noticed afterwards that the 120uF primary smoothing caps installed (wired in parallel) were only rated at 200 V rather than the 400 V required when running at 230, and as specified in the Sam's schematic. Those have now been replaced with 2x100uF parts from Jaycar and the PS is back in the monitor ready to test the one remaining unknown, the degauss circuit's PTC. Just switched it on and so far all is good, no smoke.

It's clear that Tatung designed the monitor for world-wide voltages but took a few liberties in production on 120 V versions. Three EMI cross-line caps also needed upgrading to 250 VAC parts and I hope there is nothing I've missed. It also ocurred to me that when Sams writes a service manual that they only measure and document what they find in the particular example they tear apart and are unable to quantify that relative to the original design specifications. In the schematic there are two hints but nothing specific that the PS can handle 230 V, where primary PS DC voltages are doubled over what they quote as test values.

Modem7, if you have one of those plug-in power monitors as a last check perhaps I can get you to measure the load on 230 V ?

At the currently-supplied 240 V it's reading 0.40 A (.59 PF) and 56 watts with the text screen shown.

Lastly, I've changed the fuse from 4T to 2T as it will provide better protection at 230.

IMG_0269.jpg IMG_0265.jpg IMG_0270.jpg IMG_0271.jpg IMG_0272.jpg
 
Last edited:

modem7

Veteran Member
Joined
May 29, 2006
Messages
8,474
Location
Melbourne, Australia
Modem7, if you have one of those plug-in power monitors as a last check perhaps I can get you to measure the load on 230 V ?
I do not have such a device, but I can measure AC voltage and amperage, but not power factor.

245 Vac
------------
My 5153002 with video input (a few lines of text) = 0.43 A
My 5153002 without video input (white screen) = 0.48 A

120 Vac
------------
My 5153002 with video input (a few lines of text) = 0.55 A
My 5153002 without video input (white screen) = 0.64 A
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
Amazing how close the current is at 120 VAC compared to 245 VAC ... and considering the label on mine quotes 0.95 A @ 120 VAC. Looks like I'm good however regarding power supply efficiency and thanks for taking those useful measurements.

After a decade my (2) 5150, (2) 5151 and (1) 5153 are finally all converted from 120 VAC to a region-appropriate 230 VAC, without substantially affecting their vintage appearance. Never thought I would see the day.
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
Down to the design of the power supply, I expect.
The current waveform below (sourced from a crude clamp meter) is what I see at 245 Vac. Peak is about 1A.

Well, what I mean is that since the primary DC voltage is doubled to 306 V (roughly) the input RMS current draw should in theory be half since the power delivered is the same.

I can only assume the current overshoot is the result of diode recovery time but I'm not sure what causes the current to subsequently ramp slowly back?

My impression is that Tatung added this separate enclosed power supply to an existing monitor design to satisfy a world-wide line voltage requirement from IBM. Clues are that there is an EMI filter directly following the mains input then another inside the PS, along with an internal fuse rather than a chassis fuse. And the PS outputs 115 VDC, fed back to the chassis near the mains input.

It's also interesting that this PS uses only discrete components - no ICs as found in the earlier 5150 63.5 watt power supply. That probably reflects the technology and design practices present in the US vs Taiwan at the time.

I just measured the current at 235 VAC running the monitor via a step-down autotransformer and I'm getting 0.29 amps / .74 PF / 52 watts vs 0.40 / .59 PF / 56 watts running directly off the mains. I can feel it does run hotter at the top surface so I'm going to see if I can track down the reason for that extra 4 watts, ignoring the added minimal loss from the transformer.
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
... I'm going to see if I can track down the reason for that extra 4 watts...

I've bench-tested this power supply on both 110VAC and 230VAC and found one resistor that is running significantly hotter at the higher line voltage. Since I know nothing more about this design other than that it's a "ringing choke converter" design, and don't have the test equipment to evaluate design changes, I'm going to have to settle for simply improving the heat transfer of the resistor to the metal enclosure and crossing my fingers.

Perhaps one day someone delving inside their 230VAC 5153 might compare and document primary-side component values relative to the Sams schematic, which was derived off the 110VAC model.

R805%20heat%20issue_zpsudchebyk.png
 

Attachments

  • IMG_0327.jpg
    IMG_0327.jpg
    92.1 KB · Views: 1

Malc

Veteran Member
Joined
Jul 16, 2009
Messages
2,223
Location
UK
I just had a quick look inside an old 230V 5153 PSU i have which has suffered heat damage to the PCB and R805 is 5W 30 Ohm
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
Great info, thanks! Interesting that it's exactly doubled. I thought of changing it but was unsure which direction to go, or if I might damage something.

This handily negates my theory that the power supplies are exactly the same and I'm lucky that mine has not gone up in flames. I would expect that there are other minor differences in resistor and cap values in the area between R805 and T802, but the circuit design is likely the same.

As best as I understand, the circuit consisting of the upper left coil of T802, via R805 and C821 provides positive feedback to the switching transistor Q801, encouraging it to oscillate, modulated by a voltage regulation circuit to the left and passing through the overload crowbar circuit to the right.
 

Attachments

  • IMG_0329.jpg
    IMG_0329.jpg
    95.4 KB · Views: 1

Malc

Veteran Member
Joined
Jul 16, 2009
Messages
2,223
Location
UK
I just got a Pic of mine, Not very good, crappy old camera. Little difference in layout to yours.

IMG_3376.JPG
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
Compared side by side the board layout has some differences. If the p/n is what it appears, the same, yours is a later revision, -2.

a) the transformer is the same p/n (phew!) but is it's core is turned 90 degrees.
b) the enclosure has more holes, especially at the top (right in photo.) You can see both suffer from high temperatures.
c) the lower connectors and 10 watt inrush current limiting resistor are relocated.
d) the power switch circuit has another set of conductors and 2 more connector contacts (lower right.) The wires now run so close to R805 they had to install a insulating tube over the resistor.
e) there's a green wire exiting to the left, presumably a ground wire for the blue Y2 EMI caps, which look a bit larger as well, perhaps the reason for the change.

Three large electrolytics on mine are smaller because they've been replaced. It's anyone's guess if the schematics are identical or revisions have been made to the design.

Thanks for taking the pic, Malc!

attachment.php
 

Attachments

  • combined 110-230.jpg
    combined 110-230.jpg
    106.2 KB · Views: 1

Malc

Veteran Member
Joined
Jul 16, 2009
Messages
2,223
Location
UK
It wouldn't surprise me, Some manufactures made multiple revisions, One day i'll remove the PCB and have a good close look and compare to the schematic, It needs repairing anyway.
 

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
Not always an easy task but hopefully the underside will have component labels.

In fact, if you can take the photo I'll make a schematic.

attachment.php
 

Attachments

  • 012.jpg
    012.jpg
    101.2 KB · Views: 1
Last edited:

paul

Veteran Member
Joined
Mar 18, 2004
Messages
809
Location
New Zealand
To update this thread, I've given up on the 5153's antique and inefficient power supply. Even though I got it to work on 230VAC, it runs warmer than I'm comfortable with.

Standard industrial power supplies are nearly impossible to find with a suitable output voltage and power rating but some came up on eBay a few weeks ago. There are seven left if anyone else is contemplating this moderately-difficult upgrade. The unit is of adequate quality, the price is right, the output is clean and stable and can be easily adjusted from the specified nominal 110VDC up to the ideal 115VDC. It's rated at 100W and the 5153 needs roughly 60W. It has an internal switch for 100-120/200-240 VAC input.

Aside from implementing mounting and wiring, the original degauss circuit (just one component, a PTC resistor) needs to be cut from a corner of the old PCB and mounted somewhere where it can dissipate heat (first pic at the top.) Also a main fuse should be added just after the monitor's power input.

P1010858.jpg P1010859.jpg
 

andrew96

Member
Joined
Oct 21, 2017
Messages
31
Location
Kent, UK
Compared side by side the board layout has some differences. If the p/n is what it appears, the same, yours is a later revision, -2.

a) the transformer is the same p/n (phew!) but is it's core is turned 90 degrees.
b) the enclosure has more holes, especially at the top (right in photo.) You can see both suffer from high temperatures.
c) the lower connectors and 10 watt inrush current limiting resistor are relocated.
d) the power switch circuit has another set of conductors and 2 more connector contacts (lower right.) The wires now run so close to R805 they had to install a insulating tube over the resistor.
e) there's a green wire exiting to the left, presumably a ground wire for the blue Y2 EMI caps, which look a bit larger as well, perhaps the reason for the change.

Three large electrolytics on mine are smaller because they've been replaced. It's anyone's guess if the schematics are identical or revisions have been made to the design.

Thanks for taking the pic, Malc!

attachment.php


I have the psu on the right, the TPW246 transformer IS different to the one on the left, it is marked B after the model and has less connection pins from the earlier version.

the earth wire is from the isolated secondary 0v and is just connected to the chassis on the right psu... the right psu isin my 5153002 100 to 250v marked on the back


Now I bought my 515302 with a dead psu, found the output diode D815 short and the main chopper transistor 2SD1185 short and fuse open as well as R801 3r3 10w.

replaced them all and the capacitors too! for saftey ran it up from a current limited dc power supply on its input and a 60w bulb on its output, it functioned perfect, stable 110v output with 100 to 240vdc input!! great!!!

IMG_8283.jpg

I next connected the psu to the monitor and again powered it from the voltage and current limited dc psu, worked fine!!!!

IMG_8288.jpg

so the next move plug it into 240v ac mains!! , after all this work as a BANG!!!! now the psu is a right mess! bits blown that are unidentifiable!! discovered the 2SD1185 I fitted was a fake! so that's why it failed when full surge voltage was applied!!

so now I have to identify all the blown bits and rebuild it again from far worse than before!
 

andrew96

Member
Joined
Oct 21, 2017
Messages
31
Location
Kent, UK
So basically it took out D817, R821, R809, R823, D820, Q802 2SB739 and of course Q801 2SD1185!

it is a Tatung PWB-811 psu


IMG_1302.jpg

after alot of searching I found this!! very similar!!!

2011410215951178.jpg

but no component values!!! :(

can I find anything about a PWB-811 psu? nope, the whole of the Internet and nothing!!

so back to square one!!!!

seems to be just a UK model!
 

andrew96

Member
Joined
Oct 21, 2017
Messages
31
Location
Kent, UK
So these are the bits I need help with!! anyone know there values??

IMG_1303.jpg

top diode marked TSR207, no info on the internet! has to be a high speed diode around 2A at 700-1000v I think!!

2 gray resistors marked 1100 and 1229, look wire wound but markings don't make sense to any resistance values!

one blown carbon resistor, quite sure this is 62r 5% can just make out the coloured bands!!!

2 diodes, one marked T3, one marked with nothing! both do-35 case and black band

transistors I now have (non fake ones this time!)

anyone help???
thanks in advance
Andrew
 
Top