Re: 547 Recap

Harvey White

I can't help you with some of these things, but perhaps I can tell you what you're doing in some of this.

The deflection plates of a CRT should (for various reasons) be driven in push pull.  That is, if you have a signal at the input of the amplifier, it should produce two signals, one going plus, one going minus.  This gives you effectively twice the sensitivity of the case where you hold one plate fixed, and feed a signal to the other.

Somewhere in the horizontal or vertical amplifier chain, you have a single signal.  There's a circuit that produces a positive going and a negative going signal from one signal.  The vertical or horizontal amplifier amplifies each signal and drives each plate from that resultant signal.  You have one chain doing a positive signal, one doing a negative signal.

If you were to short the deflection plates of the CRT, then if the geometry is right, the spot ought to be in the middle (either H or V) depending on what you shorted.

A useful technique is to go from the plates back and short the equivalent signal - to +.  If you don't get a display where the dot is exactly in the middle, then the amplifier is not giving you the same signal + and -, and you may have a bad component.

Anything you get where the signal is synchronous to the line generally indicates either ac line leakage to the amplifier, or bad power supply ripple.

Shorting two  transistors' collectors in this differential amplifier doesn't really increase the load, it just makes the + and - channel signals identical.

Proper working of a scope says that not only are the power supply voltages within tolerance, but the ripple is, as well.

Nasty things happen with ripple.


On 5/9/2021 7:42 PM, J via wrote:
I have a 547 that I'm thinking about recapping. But I have a few questions before I start working on this, as I was having trouble finding info in the search function on the website.

I was attempting to recalibrate the scope when I got to step 9d. on the vertical amplifier: "Short the cases of Q1074 and Q1084 (see figure 5-4) together (not to ground) and note the trace deviation from electrical center. Deviation should not be more than 0.5cm maximum. Remove the short."

When I performed this step, I had about 7cm of deflection peak-to-peak (about 3 to 4cm from electrical center), and was asynchronous to the line. I started suspecting that it was a problem with the power supply, and I measured the peak-to-peak voltage to be about 200mV on the +100V rail when the collectors of these transistors were short. I reasoned that shorting the transistor collectors increased the load, and caused additional noise on the power supplies. And as a result, I found 200mV ripple on the power supply spec (exceeding the 15mV specification), so I need to at least find the culprit capacitors, and possibly recap the unit.

I then started looking at the locations of all of the power supply caps within the 547 and realized the mounted ones generally have more than one internal cap, and were mounted in ways that were very difficult (impossible?) to remove the mounted capacitors - under the parts on the ceramic terminal strops. The mounted capacitors were all 35mm in diameter.

* If I order caps that are still 35mm in diameter, can I use the same mounting fixtures?
* How do I gain access to all these capacitors without disassembling the entire housing?
* If I recap the timing caps while I'm doing this - what voltage rating should I use? It's not specified in the parts list.
* I'm counting about 15 capacitors with lethal voltages on them - not sure I want them dangling around in the chassis. How have others gone about mounting them securely?
* I've thought about just leaving the original caps and adding the new caps where I can find good locations in the chassis. This isn't very clean though, but I'm still curious if others have had success with this approach.


Join to automatically receive all group messages.