H89 Video Board Troubleshooting

[Matthew Feurtado]

Greetings,

I recently acquired a H89 and am in the process of getting it up and running again. So far I have the CPU board repaired, couple dead 4116, tantalums, and a 7905. Tested the CPU board with a Wyse terminal. On the Terminal board I have replaces some more tantalums and it produces the expected startup beep as well as H sync, V sync, and a video signal. So I think that is all working well but that brings me to the Video board that is putting up a fight and I am not sure what is wrong.

With everything connected and the Terminal board wired up for test I get no picture but can hear the coils hum and can't see any glow from the filament. The +53v supply drops down to around +33V when it is powered on like this. If I disconnect the flyback transformer at connector P203 then the board powers up with +53V spot on.

I have not yet replaced any components on the Video board yet. I tried measuring the flyback windings and the low-voltage secondary winding  (yellow to blue) read 0.8 ohms while the one schematic I found that was annotated shows this should be 1.4ohms. My only other clue is that when the flyback transformer is connected the heat sink gets very hot, like burning finger hot, so maybe the issue is with Q214 the HOT transistor?

Any pointers or tests to try would be appreciated!

Thanks,

Matthew

[Joseph Travis]

Greetings Matthew,

Congrats on getting as far as you have!  I know from experience that troubleshooting a H89 with multiple problems can be very trying.  I was a Computer Tech for Heathkit in the 80's and repaired many over the years.  It sounds as though you are very close to fixing this.  I checked the resistance of  the YEL-BLU wires of the flyback in one of my H89s and it read 1.5 ohms, RED-ORG is 3.9 ohms, BRN-BRN is about 3 ohms.

In all the time I serviced computers, terminals, TV's and monitors, I never encountered a bad flyback (except a bad diode stack in the anode lead).  That's not to say it couldn't happen though.  You should be able to check Q214 using a multimeter with the flyback disconnected and NO POWER applied.  If it turns out to be the flyback, I believe someone in the group may be able to salvage one from a video board that has a burn hole in it.

Good luck!

Regards,

Joe

--
You received this message because you are subscribed to the Google Groups "SEBHC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sebhc+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/47c8c7b8-99ba-49d1-b206-e642f10e6e67n%40googlegroups.com.

[Matthew Feurtado]

Thanks Joe,

I am so excited to have found one locally. The H89 is from well before my time but I do have a connection from a story my Grandfather told me about the time he almost bought one when he read a add in the paper. When he showed up at the address it was a hippy selling them from his driveway and my Grandfather decided against it at the time. I believe he ended up getting either a TI99 or a C64 instead. 

Any ways I measured my flyback again and got the following. 

Yellow-Blue = 0.4

Red-Orange = 1.3

Brown-Brown = 0.1

and infinite between Black and all others and between yellow and red.

This is for part number 51-200, some other versions of the schematic show 51-196. Not sure if they would have changed the part number between rev if the part itself is the same.

My measurements aren't the same value as yours Joe but given the low value and that the ratios are pretty much the same between primary and secondary I'm not ready to write it off as bad just yet.

I will try Q214 next and then maybe order some new caps just to see if that helps.

If i could get a hold of a spare flyback that would be the easiest part to replace since it just plugs in. If anyone knows where I could get one let me know.

I appreciate the help to bring this beautiful machine back to life.

Best regards,

Matthew

To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/CAGQDgBDFx4E95roT5qdM73Ww%3D-%2BgOHsLM8bdo6514U6z-NNGNQ%40mail.gmail.com.

[Glenn Roberts]

Matthew: have you re-capped the large capacitors in the main power supply? That’s usually a first step to ensure good, clean power. I’ve observed a range of issues getting magically fixed after cleaning up the power…

Sent from my iPad

On May 17, 2023, at 11:47 PM, Matthew Feurtado <mwfeu...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/CA%2BsP3JS24HTxAFVFxCDASievCY%3DcJfV_9wbgjFToNZmONkJPWw%40mail.gmail.com.

[Apropos Sales]

Have you tested the Q214 (or 217) horizontal output transistor and flyback diode D208.? I had a problem with a horribly mouse-contaminated video board. The H output transistor showed C-E low impedance in circuit. I removed the TO3 device, only to measure it again and find a normal (open) reading. I replaced the device and after a fraction of a second, the C-E showed shorted. Thinking it may be some breakdown caused by CRT arcing, I applied high voltage dope and grease, etc., to no effect. This occurred 3 times. It turned out to be that the mica insulator had a hole in it and somehow the case made contact with the metal heatsink bracket. In my case, this caused the 53 volt regulator circuit to fail, as well as R256 1r2 2W and R211 0r51 2W.

Replacing the insulator fixed the problem. (While waiting for new insulators to arrive, I swapped the two TO3 insulators. The power supply transistor doesn't experience the high voltage flyback pulse, so it doesn't arc. (Famous last words.))  I also didn't need the replacement TO3 transistors I had purchased.

Do you have a scope? Do you have a variable DC power supply that can do 53 volts with some current? If so, you could lift R256 1r2 2W away from the 53v rail, connect the supply to the lifted end, slowly bring up the voltage, and measure the current and look at the horizontal drive and flyback. At full drive, the CRT filaments should be hot.

To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/5A92D58C-1BCE-4EF3-9B95-8FBCB6F89430%40gmail.com.

[Matthew Feurtado]

Success! Today I finally have achieved victory over the video board and have a working picture on the CRT!

The picture still needs some adjusting as the brightness has to be turned way up to get any display so there is a lot of blooming in the characters. Found there are more bad RAM but that is a much easier problem to solve than CRT drivers.

Ultimately the parts that had to be replaced were the horizontal output transistor, the HV flyback diode, a 7905, and many tantalum caps. The HOT possibly failed while I was testing hooked up with a benchtop power supply for the +53V but it eventually measured bad with a multimeter. The flyback diode was the final piece of the puzzle. Through process of elimination I determined that the onboard +53V supply was stable in any test configuration except with the HV lead connected to the CRT. With that disconnected and floating the +53V was solid and even the filament started to glow brightly. With it connected the +53V fell to around +30V. Luckily that diode is not potted inside the flyback so replacing it was pretty straight forward. No idea what the original manufacture was it is marked as "A18-20 14" which turned up no leads. The annotated schematics show a HV of 15kV so I went looking for 20kV diodes. I did check Digikey first but decided to go with some 20kV 20mA I found on Amazon. I couldn't believe my eye when I saw these had a 5 year old review where someone repaired a H-19 terminal with these exact same diodes, the stars aligned. Amazon link is https://www.amazon.com/dp/B074Y9DXDB?psc=1&ref=ppx_yo2ov_dt_b_product_details Best I can tell the manufacturer of these is probably HVCA or HVCAPS going off the 2CLG in the title.

Funny enough I didn't realize it until I cut open the heatshrink around the old diode but there was a bulge in the middle where I assume the old diode had cracked and let out some magic smoke at some point creating a air pocket under the heat shrink.

For those interested the horizontal output transistor I used to replace the failed Q214 is the NTE238 available to order on Digikey from NTE. https://www.digikey.com/en/products/detail/nte-electronics-inc/NTE238/11650001

Anyways thanks for the pointers I am pleased to see this machine come to life.

I will be back when I have replaced all the bad RAM and sorted out display adjustments. At some point I will want to do more than just play with the "H:" prompt :)

Best regards,

Matthew

To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/CA%2B8v2v2Ge6kxoXNeJaRQ_duSD%2BkWh75SnPZyziTn6PifdDGaVg%40mail.gmail.com.

[Glenn Roberts]

Great electronics detective work and a great outcome. Congrats! Always wonderful to see another old Heathkit come back to life!

Sent from my iPad

On Jun 23, 2023, at 12:49 AM, Matthew Feurtado <mwfeu...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/CA%2BsP3JQTDgoj05i%2B%3DxwFHts9FdQ0w6USQGPHVyaY%2Bqu_Ui5grg%40mail.gmail.com.

[Joseph Travis]

Congrats!

To view this discussion on the web visit https://groups.google.com/d/msgid/sebhc/9BD189A0-4DC1-47D0-9FEE-C47A472882C8%40gmail.com.