Tektronix 49x Spectrum Analyzer Deflection Amplifier smoking.

[Gazucha]

 Hola again,

 This is actually an extension of an on-going project which was featured in this post...
 https://www.eevblog.com/forum/projects/tektronix-49xx-spectrum-analyzer-series-power-supply-failure/

 Due to the fact that that particular issue was resolved, it seemed appropriate to start a new thread for this latest hurdle.

 In a nutshell, I had repaired a Tek 492 PSU yet when it was reconnected and turned on, we had no display and the Deflection Amplifier started smoking from the 800K resistor in photo 2 below, at which point the Tek 492 was switched off and the offending module removed.

Although the amplifier is reasonably clean, after a closer look it was clear that there had been some overheating in a few places. Most notably one of the LM318JG's. Photo 1. There are 6 of these ( 3 pairs) and the one shown is the only one obviously damaged. The smoking resistor still reads 802K and the other resistors which appeared heated are all reading fine too.
 
 In photo 3 you can see the two LF13333's which don't look best soldered.

 So has anyone had this problem before? Is there common prob's with these modules?
 And considering as the instrument has to be re-calibrated anyway, how important is it that I find an identical LM318(JG)?

 As I am many miles away from the Tek 492, it would be good to resolve this properly before I head over to plug it back in.


 And as always, any help would be bloody marvelous.

[Orange]

It might be a short in the output transistors...

Mind you that these transistors are not having the usual pin-out. I have successfully replaced them with Japanese types, since the originals are unobtainable.

Tektronix used to replace the complete module. Also there is a good read on the resistors from John here http://www.ke5fx.com/49x_notes.pdf

[Gazucha]

Hi Orange,

thanks for the reply and especially the replacement transistor information.

I have the resistor article, so checked those mentioned in it and they measure ok, but they were for display noise/jitter problems whereas at the moment we have no display at all.

It did cross my mind about the transistors, so I'll have a lookee-see before I head off to buy an LM318 or two. I mean, something caused the LM318 to burn, although considering the Tek 492 has been sitting for a number of years, it could perhaps have started failing through a dry joint?

To edit... output transistors are ok...

[madsci]

No display at all... Do you have the horizontal? Do you have HV?

Loss a vertical usually results in a bright horizontal line unless the system can detect the loss of vertical and shut down.

[Gazucha]

Hola mate,

 to be honest, it was tested in a sunny yard, but we could see no light on the display at all. Am returning on Monday and will put it indoors to see if we get at least a glow.

 The situation, is that this unit is located 2 hours travel (each way) and I have to transport an inverter and transformer each time, just to test it! The owner wants it sold, however it was totally dead and being a complex instrument no-body had wanted to buy it, so I offered to try to get it to at least turn on.
 That much we did, but within thirty seconds an 800K resistor on the Deflection Amp (located at the base of the CRT) started smoking, so we switched it off and pulled the board. No time to test any internal voltages.
 The only damaged component I found on the Defl' Amp was the LM318, so that was changed and all poor solder joints were touched up ready for the Defl' Amp to be reinstalled.
 
 My basic - and very rusty - understanding of CRT technology (learned 30 years ago at college) left me thinking that perhaps the CRT display needed the Defl' Amp signal to generate anything at all but I have been reading up and now have a clearer understanding of CRT systems.
 
 Yesterday I also found the PSU diagram with the correct output voltages (which I didn't have before) so will check these first before I reconnect the Amp and turn the instrument back on. Perhaps I was a bit carefree in assuming all the PSU output voltages would be good, after the probe connection pin readings came back spot-on (+15v -15V +5v). Perhaps one of the output voltages is now too high?

  There is also a small HV board which sits alongside the CRT at 90 degrees to the Defl' Amp, although I am not too sure as to how to go about testing this. I am reading the relevant part of the manual today in order to clarify this.

 We have to remember this unit has been sat in a dusty storeroom for 7 years and who knows when it was last used before that? It is in good condition. Super clean inside. Just needs some tlc.

 Any input you can give will be well appreciated.

 Cheers,

 Gazucha.