• Please review our updated Terms and Rules here
  • From now on we will require that a prefix is set for any items in the sales area. We have created regions and locations for this. We also require that you select a delivery option before posting your listing. This will hopefully help us streamline the things that get listed for sales here and help local people better advertise their items, especially for local only sales. New sales rules are also coming, so stay tuned.

Wanted: Power Supply Repairs.

jhuyett

Member
Joined
Jan 13, 2017
Messages
34
Location
PA, US
Hello All,
Looking to see if anyone is willing and able to do repairs on old power supplies.
I have two complete PS's for a VT-100 (both with different issues) and a 3rd for parts.
Also, a PS for a Bernoulli 10+10. This I might be able to replace with a new mini/ micro PS, but haven't verified and would prefer to be able to use the original.

I've done basic soldering/ troubleshooting before, but not on anything like this.
I'm in the Lancaster area of PA US.

Regards.
 
According to here:

http://www.vcfed.org/forum/showthread.php?38234-Fixing-a-VT100-power-supply
https://web.archive.org/web/2014091...2013/07/19/building-a-new-vt100-power-supply/
https://web.archive.org/web/2014091...-power-supply-for-a-DEC-VT100-serial-terminal
http://www.datormuseum.se/peripherals/terminals/vt100

VT-100 documentation:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/terminal/vt100/

Seems like the PSU is known to have reliability issues, which is no real surprise since it is 42 years old. It unfortunately uses some proprietary components that don't have replacements or equivalents available, so repairing it may not be possible.

Switching supplies usually aren't too difficult to troubleshoot, you just have to work your way from the mains side towards the DC side. I'm sure there are going to be bad components that need to be replaced, just follow the schematics and test components. I would recommend using a variac though and a current limiter, like an incandescent lightbulb in series with the supply. Slapping full mains voltage across an old unknown condition piece of gear is not a great idea and can result in irreplaceable parts being destroyed.
 
Hello GiGaBiTe and thanks for the response.
One week later: I saw this come in then completely forgot. Apologies for that.

The most I've done in the past was to look for blown components or check continuity.
I had found the schematic MP00633_VT100_Schematic_Feb82.pdf previously and wish the labels on the board layout (p21) were clearer. I had tried to go through/ process of elimination without any luck.
One blows fuses and the other doesn't have any voltages on the -23v rail. I am motivated again to give the "-23v rail" another shot.

Still open to anyone willing and able to take a look at these.
 
For the one with the missing -23v rail:

According to page 27 of the schematic, there's really not much between the output of the -23v rail and the source transformer.

The only component between the transformer and the -23v output is a 300 ohm 10W resistor. There are also three capacitors to ground (C8, C18 and C37), and finally a rectifier diode on ground ( D28 ). With so few components, there's very little that could be wrong on this rail. The first two things I'd suspect are the 300 ohm 10W resistor being open, or the rectifier diode being shorted. If those two components turned out OK, I'd then check the three capacitors to see if they were either shorted, or are electrically leaky or have high ESR.

If all of those components test OK, then it would mean unfortunately the winding on the transformer is open. There's really nothing you could do to fix that, unless you have experience rewinding multitap switching transformers. You can test for this with the power supply off and make sure the line capacitors are discharged. Use ohms mode on your multimeter and see if you get a reading between the -23v tap and ground. If you get nothing, then the winding is open.

If the winding did blow open, then there's a pretty serious fault somewhere in the terminal, I would check resistance to ground on the -23v rail from the power supply connector on the edge connector without the power supply installed and see if you get a very low reading or near 0 ohms, indicating a short.

For the one that blows fuses:

Blowing fuses is almost always a fault with the primary side of the power supply. I would check the bridge rectifier first and see if one of the diodes is shorted (D16, D17, D18 and D37). In diode mode with a multimeter, you should get a voltage drop between 0.5-0.8v when measuring from the cathode (-) to the anode (+) and OL or nothing when reversing the test leads. If you get continuity in both directions, or a dead short, then the diode is bad and needs to be replaced. You may have to lift a leg of the diode to do a proper test, because other components in circuit with it can give you erroneous readings.

The next thing I'd check are the line capacitors (C9 and C14), you'll need a capacitor tester for this. I'd recommend one of the cheapy Ebay ones because they're good enough for stuff like this. https://www.ebay.com/itm/353404276114

For capacitance, you should get within +/- 20% of the capacitor rating, but obviously the closer to the rated value the better. If you get a reading too low, it generally means that the capacitor has dried up. If you get a reading too high, it usually means that the capacitor has become electrically leaky. In either case, it needs to be replaced. When a capacitor gets real bad, it can short or go open circuit, the former of which can cause the fuse to blow. I have had shorted line capacitors in power supplies before and they do cause headaches.

The last major culprit are the 100k ohm 1W resistors (R19 and R20). These can drift in value over time, short or go open circuit. If the former two cases happen, they can cause the fuse to blow. It would generally be pretty obvious if the resistors were a problem because they'll cause the PCB under them to discolor.

If none of these are the problem, I'd start checking the switching circuitry feeding the second tap on the transformer on the primary side, which is on the upper left of page 27, under all of the components I just mentioned. There are a lot of diodes and a few transistors here that should be checked, along with the couple of resistors.
 
Last edited:
Man, I just looked at the schematic of the VT-100 power supply and have to say that has got to be one of the difficult to work on and overcomplicated power supplies I have seen. An I have worked on many switching supplies. First off the whole idea of how they approach the input voltage for 115/230 and the crazy idea about how they are using a voltage doubler in 120 volt and a full wave bridge in 230, how T3 uses the emitter current of Q7 to trigger Q7 with C19 providing the first start cycle but what the hell is the point of D19?
Long story short is that the primary side of the power supply is some weird Armstrong oscillator with no regulation or control whose only advantage was that they were able to cut cost by not having to use a conventional power transformer unlike modern switching supplies that use feedback to provide regulation.
Looks like the power supply is such a pig that it cannot start osculation on the primary side without the help of the little +12 volt supply because the inrush current would be so large that the primary oscillator would be swamped and fail to start.
Think this power supply is a great example everything that was wrong with switching supplies before the technology matured to what it is today.
But then again to anyone who has ever worked on the power supplies on the Unibus PDP-11 or some of the other DEC equipment DEC can be credited with creating and using some of the strangest and overcomplicated power supplies ever built.
In terms of trying to help, cannot offer much beyond checking component value, the capacitors are somewhat important C9 and C14 and past experience have seen them short. The diode bridge D16, D17, D18 and D37 have been know to short and q7 the switching transistor can short. Any of those items will cause the fuse to blow as soon as you turn the power supply on.
The secondary side of the power supply is fairly straight forward with the -12 and -23 volt supplies being Dog simple, the thing to remember about them is if for some weird reason the rectifiers fail D28 and D29 they are not off the shelf diodes but have to be replaced with high speed diodes, have seen several people make that mistake with other switching supplies.
Same thing with the +12 volt supply and D36, its also useful to have a scope and look at the ripple voltage on the multipole DC output buses. Working power supplies should have less then 0.1 Vpp on any output leg. Also, a scope will allow you to see if the primary side is running but beware! The primary side of the power supply is at AC line potential. DO NOT CONNECT OR TOUCH ANY SIDE OF THE PRIMARY SIDE WITHOUT USING A ISOLATION TRANSFORMER! The results of working on AC Hot inputs of the power supply can be lethal if you don’t know what you are doing.
Also I would be careful about using a autotransformer and gradually increasing primary voltage being some devices on the primary tend to draw more current in undervoltage conditions, and autotransformers are not isolation transformers.
Think the thing is to view the primary side of this type of a power supply as a flywheel, the flywheel is kicked into spinning by C19 and that provides just a small kick that starts Q7, T1 and T3 into osculation (spinning), once the wheel spins up the secondary side draws energy from that wheel to drive the display. Because the wheel would never start to spin with the drag of the secondary supplies they have the little 12 volt starter supply that removes the load from the secondary side until the wheel up to speed. Sounds like a long time but its only one cycle in reality.
I know this has been a long and complicated and maybe not very useful explanation but sometimes find just reviewing the basics of switching supplies is worth it.

Good luck, Qbus
 
Back
Top