HeathKit IO-104 Repair - No Trace

[mentaldemise]

I got a Heathkit IO-104 today for pretty cheap. The guy I bought it from said the trace was present for about 20 minutes and then went away. I have the original book that came with the kit and a print of the diagram. I went through it and checked the voltages and they all seem to be within tolerances. Is there a way that I can tell if the tube itself is bad? As I mentioned the voltages that the book said to test are all within range and the calibration circuit is outputting all the signals and voltages it should be. It feels like the CRT heater is running but I'm afraid to get too close to it since 2KV will hurt pretty bad. My meter did read the HV lines as well and they were within spec.

Circuit Diagram: http://www.heathkit.nu/IO-104_schematic_4096.png

[james_s]

Turn off the light and look at the neck of the tube, you should be able to see the heater glow. Did you check all the voltages at the CRT socket? You need the right voltages on the cathode, grids and accelerating anodes to get a beam on the screen. Also I would look at the Z blanking circuit that blanks the beam during the retrace between sweeps. You do need to be careful of the HV, but it's not going to shoot a lightning bolt across the room and fry you, just be careful where you poke your fingers.

[tautech]

Sounds like a temp related fault and you might be able to isolate it with freeze spray.
With CRO's one area that's under stress is the HV supply so check HV caps and resistive dividers.
Most HV diodes have a high Vf, often 10V+ but are low current capable so be mindful not to subject them to much current when improvising some checks.
Tube failures are rare, if it goes fine for a short while it will be something else.

[tautech]

This can be handy:
http://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf

Note the voltage measurements on the schematic within circles. For these to be applicable the scope is normally set in some default state, check the manual carefully for those instructions.

[mentaldemise]

The orange glow is present at the tube. Pin 3 for blanking is putting out -175V. In my scematic that section is circled and they wronte "1480V" next to it. If I'm reading this correctly I should expect to see 68V at the base of 402 and somewhere close to that on the Zenner near it. I'm getting ~12 Volts out of the 10K 10W resistor that feeds this. 192 comes into that resistor and 12.1 comes out.

[mentaldemise]

Side note: 1480V stings a little.

[mentaldemise]

Did some more testing along with shocking myself and HH is actually only at 2.5V instead of 5.  Also the measurement at R418 is 12V instead of ~68V

I may be in way over my head. Am I correct in thinking that something on the other side of the resistor dividers is pulling too much current and dropping the voltage? The input voltages all seem to be correct from the transformer.

[tautech]

Did some more testing along with shocking myself and HH is actually only at 2.5V instead of 5.  Also the measurement at R418 is 12V instead of ~68V

I may be in way over my head. Am I correct in thinking that something on the other side of the resistor dividers is pulling too much current and dropping the voltage? The input voltages all seem to be correct from the transformer.

That's what I would suspect. A look over the schematic shows very few caps between the rails and Gnd excepting near the PSU. Then check transistors that offer a direct path to Gnd if they are shorted, use the one diode drop check. 
There's a # of high wattage resistors in series on the rails, check that none have fried and gone OC.

[romhunter]

Side note: 1480V stings a little.

I'd say I laugh a little. Tube equiment are quite sturdy IMO, so good luck trying to fix it. I'd go with the advice tautech gives.

[mentaldemise]

I ordered some new 10W 10K resistors as I did find one that is out of value when measured in circuit. I swear it measured 6M at one point and then 5K another. That's the one that 12V is coming out of that should be 68V. The 5W resistor that's putting out 2.5V instead of 5(4.7) tested correctly so I'm guessing there's something else on that board that's breaking it.

[tautech]

If you use the setup outlined in Note #6 and check any voltages circled that are sourced from +5V you might be able to identify an area that has a lower reading and therefore signify where needs looking at in more detail.

We don't know (can't see) how tightly packed the boards are or how accessible they are so you might have to rip them out and get them on the bench to go much further. Then you'll need a bench PSU but be sure to current limit it with resistors as the scope PSU does.
Sometimes the razor blade trick can be easier than lifting component legs to isolate areas and work through the 5V areas.
Can you see and parts of the boards that have run hotter than others ?

[mentaldemise]

Really appreciate the help and suggestions! The power board is easy to get to, the other boards are not. There's one on the other side of the unit as well, but the 5V rail that's at 2.1-2.5 is going to the bottom board in the picture. It comes in by the metal can'd cap in the back and then through a Zener. I pulled that zener and it tested correctly. The cap is in the 4th picture. You can also see some of the short leads that make disassembly a nightmare. The top-right most resistor is the one that should be giving me 68V and is giving me 12 instead. That's the blanking control circuit also. The white wire coming out goes to pin 3 of the CRT.

The second from-left one on the bottom(5W120Ohm) is the one that has 2.5V on the bottom and 18 on the top but should be 5 on the bottom and 18 on the top.


As for heat, the whole power board gets hot, all the resistors are hot to the touch, not warm.

https://photos.google.com/share/AF1QipMy3wU2OtTo8OmIsZ-Zv51V2zuHtaozEmXVDqO5AQ0vZqf27a7r0EcOJa2C3EdQsw?key=QVNZOTlPY2k3QjhrT0p2aG9lQlhkd0VfbFhwaDZB

[tautech]

As for heat, the whole power board gets hot, all the resistors are hot to the touch, not warm.

You misunderstood.

Look at the underside of PCB's, there will be some areas that understandably show signs of heating (discolouration) and other areas where there's no good reason why there is discolouration. See any, hunt out why.

[mentaldemise]

Sorry. I can see the underside of those two board by taking the screws out of the PSU one and taking the bottom off. There's no signs of anything getting hot. The instructions have you mount everything off the board also though.

[tautech]

Sorry. I can see the underside of those two board by taking the screws out of the PSU one and taking the bottom off. There's no signs of anything getting hot. The instructions have you mount everything off the board also though.

Great, it's just another thing to look for that helps finding a faulty portion. 

Don't think that standing off components will save the board from being subjected to heating, leads help disperse heat and traces can also be subjected to currents beyond design by faulty components.

[mentaldemise]

I think I have all the tools to fix this thing. It's older than I am but I would still love to get it working just for preservation's sake. That includes a Siglent scope!

It'll be a few days before I get those resistors from China at least. Would hooking up a bench supply with 5V to the chassis and the output where it's supposed to be 5V be a bad idea? I was thinking give it 5V and let whatever is bad blow up or smoke to make it obvious.

[tautech]

Would hooking up a bench supply with 5V to the chassis and the output where it's supposed to be 5V be a bad idea? I was thinking give it 5V and let whatever is bad blow up or smoke to make it obvious.

Think about it........+5V is supplied via a current limiting resistor and without that you could smoke traces. 
I'd rather isolate areas using cut traces and repair with a solder bridge later.
You'll need to give some thought as to which areas you isolate maybe need one of the other rails supplied as well for tests to be indicative of that area being OK.

[alsetalokin4017]

What's the story here?

[mentaldemise]

I wondered the same thing. I cut it. I believe I found somewhere online where the schematic had that added as well.

[tautech]

I wondered the same thing. I cut it. I believe I found somewhere online where the schematic had that added as well.

What's the spec for the string of zeners ?
You need the grid at a more positive voltage to setup the electron beam.
They could be up to a 100V string to get the differential.

[mentaldemise]

When I measured the output from those zeners it was -125V with the common on the chassis.

[mentaldemise]

It doesn't look like any of the cross references i"m coming up with have the part listed with an alternative. R4507-5 (#56-79)

[tautech]

When I measured the output from those zeners it was -125V with the common on the chassis.

The measurement need be emitter Q404 to terminal A or pin 3 on the CRT

[mentaldemise]

Sorry, didn't read my meter correctly. The first Zener in that chain gets 12V in and then produces negligable output. What I said was -125V is actually -.125 mV. The chain of zenner diodes is supposed to start off with 68V.

[mentaldemise]

Sorry, didn't read my meter correctly. The first Zener in that chain gets 12V in and then produces negligable output. What I said was -125V is actually -.125 mV. The chain of zenner diodes is supposed to start off with 68V.

Q404 to A procused nearly +200V

[tautech]

Sorry, didn't read my meter correctly. The first Zener in that chain gets 12V in and then produces negligable output. What I said was -125V is actually -.125 mV. The chain of zenner diodes is supposed to start off with 68V.

Q404 to A procused nearly +200V

Which I think is in the ballpark of where it needs to be BUT understand it's a 200V difference between those 2 points as A is linked to -1400V via R5 1M .

[mentaldemise]

Great. That's one thing down. Is it possible that it's blanking all the time? It looks like the input for the blanking circuit comes from MM which is Q314's output and uses the +5 that's currently dickered. The Schmitt Trigger goes to two NPN transistors. Following things around it looks like Z should have 1.2V on it from Q314 but it appears to have 3.7V on it instead(with stability in the center and Time/CM set to 1mSec.)

[tautech]

Without the circuit description I'm not sure where blanking is performed but that's academic as until you get the +5V rail up it doesn't matter. All rails need be up and close to spec.

[mentaldemise]

Well, as it happens I have the original assembly book that came with it an includes those! https://goo.gl/photos/q3ajba1UpTk1go1D8

[mentaldemise]

I want to repeat my appreciation for your help! Am I incorrect in thinking that disconnecting M will change the voltage that I should expect there? I

[mentaldemise]

Might be way off here again. It looks like to drop 18V down to 5 with the 120 Ohm resistor they have then I'm looking at 100 mA and only need a 1.3Watt resistor.

[tautech]

I want to repeat my appreciation for your help! Am I incorrect in thinking that disconnecting M will change the voltage that I should expect there? I

If M is +5V do not disconnect while the other rails are fired up, you will fry something.
Supplying +5 from another supply is the way to go, there might be mention of the expected current draw in the manual or circuit description.
Take a step back and re-read the thread, give some thought as to next steps which should really be focussed on why the +5V rail is low.

[mentaldemise]

I was using the circuit diagram that says R407 should have +18V on one side and then goes to HH which should have 5 and then a resistor divider calculator to figure out what amperage it would draw with a 120 Ohm resistor as is in place. That gave me the 100mA. I hooked up another DC supply limited to 100mA and it drew full current. I'm guessing I have a short to ground on the 5V rail, but with 100mA nothing is actually getting hot. To actually get to things I'm going to need to disassemble the time/cm selector and de-solder its 20+ connectors.

[tautech]

Of course you need to check there's no bead tantalums on the PCB, I couldn't spot any in the schematic but if there is let them be #1 suspects.

[mentaldemise]

I don't see any either, of course searching google for "vintage tantalum capacitor" shows pictures similar to all the caps in the thing. I think I'm going to order some parts to remove some of the bodges that are in here as well. Instead of 2 X .1 @ 2000 V there are 4 .05s. There are 2 1W 500k resistors instead of a 2W 220K. I wish there was a way to get the CRT to turn on to know if I should even continue. I can't even find a replacement tube, so if that's what's wrong I'm pretty much screwed and it's a display piece.

[james_s]

I'm sure the CRT is fine, they don't tend to fail suddenly, they just get dimmer and fuzzier gradually over time. Also they're not impossible to find, you can easily sub a similar electrostatic deflection CRT.