7633 psi help, +130v supply


Mark Vincent
 

David,

C806 is bad. Replace it with a Nichicon UCY, or equal in quality, 330mfd 160V. This 130V supply is a stacked supply. The 130V supply is not grounded. Q850 regulates the negative side of this supply using the 50V as the reference. Replacing the electrolytics in the supply is a good idea.

Mark


 

On Sun, Jan 9, 2022 at 06:13 PM, David Templeton wrote:


The capacitor and rectifier is good on the main supply, A12 board. However,
I’m struggling to get my head round the 130v supply circuit. On the A11
board, the 40v line appears to regulate the 130v, but that is controlled by
the 50v supply, which in turn seems to be derived from the 130v.
In addition to what Mark writes, it may be interesting to know that the various supply voltages are linked to protect the 'scope in many cases of LVPS failure. The -50 V is the basic reference for all and stabilization is achieved by using the SENSE lines.
I recently uploaded a very good drawing of the 7603's LVPS to the 7603 article in TekScopes. It was made by member Enrique Saguid and may be found under Manuals, "Newly drawn clear version of LVPS schematic". AFAIK, the 7603/13/23/33 LVPS are (virtually) identical.

Raymond


 

On Sun, Jan 9, 2022 at 06:13 PM, David Templeton wrote:


Where does the reference and grounding for the 130v come from as both ends of
the winding are 130v and 40v lines.
Maybe it's worth noting that Q852 acts as a comparator: The base voltage is compared against the base voltage, which is about +50 V if the 130 V is correct. The voltage divider made up of R855 and R856 determines that.

Raymond


 

On Sun, Jan 9, 2022 at 10:48 PM, Raymond Domp Frank wrote:


The base voltage is compared against the base voltage,
That should have read:

"The base voltage is compared against the emitter voltage", of course.

Sorry for that, trying to do several things at the same time!

Raymond


Mark Vincent
 

David,

In my 7603, I put in: 647-UCY2D331MHD6 for 250mfd, 594-MAL225639222E3 (alt. 871-B41505A9228M000) for 1800mfd, 80-PEH536JCE5100M3 for 9600MFD, 598-81LX153M035K452 for 14,000mfd and 871-B41505A4229M000 for 18,000mfd, 78-BZX85B9V1 for VR890 and a pair of 11,000 1/2W resistors in parallel for R890. I did raise the wattage of some other resistors in the scope. If you want to know what other resistors I raised in wattage, ask and I will tell you. I do have the 7603, not 7633. I would like to find a 7633 complete. The part numbers I gave are from Mouser. These numbers are ones that are the best I could find in temperature and life. Raising the value is alright. The first number may have to have the 6 left off if it is not in stock. I unsoldered the cans and put in the new caps. Some I added short stiff wires to the lugs to be soldered in the holes as the pins did not line up with the holes. This also keeps the filters from touching the board and allowing air flow all around it. You may need to add jumper wires on the negative side of some of the negative lug holes to keep a complete circuit. Look at the traces to see if you need to do this. It is best to replace all the filters at once. You know the age and condition then. The zener is highly likely open and the resistor to it should be 1W.

If you are concerned enough about ripple current, add an NTC like 527-CL70 in the primary. Also move the fan to the 120V tap instead of the 100V. I did this in mine to get better air flow. Go ahead and oil the fan.

If you replace the rectifiers, use 4A types to replace the 1.5A types. The rated voltage is 200V for these. You can raise the voltage if you want. The current is the main thing.

Mark


Harvey White
 

That small 100 Hz dip is interesting.

Assuming you live in 50 Hz land, that suggests it's coming from the power supply.

The theory suggests that the drain on the +50 bulk supply is large enough to pull it down a little.  That suggests looking at the drain on the +50 volt line by looking at the voltage drop across the sense resistor (if any).

Another possibility  (since it's 100 Hz and not 50, I'd assume that the rectifiers are both good), is that the value of the bulk filtering capacitor is low, or the ESR is perhaps very high.  Not sure if they give you waveforms for that, but it would be something to check in passing.  IIRC, typical emitter follower pass transistors in power supplies are run with at least a 3 volt drop across them.  Might be interesting to do a differential measurement with both a meter and a scope to see if the voltage remains high enough.  With an inadequate bulk supply, the regulator circuitry might be just fine, but unable to deliver.

Just idle thoughts.

Harvey

On 1/11/2022 3:16 PM, David Templeton wrote:
The -50v is good, flat DC, as is the -50v sense feedback.

The +50v has a small 100Hz dip, the +50v sense feedback is exactly the same.

All other supplies are flat DC.

How do you post photos on to the replies?
Or do the need linked from somewhere?






Harvey White
 

Interleaved.

On 1/11/2022 3:39 PM, David Templeton wrote:
Yes, UK, so 240v 50Hz.
Figured.  Not too many places have 50Hz and 110/120.  Japan has 100 V, but 60 Hz.  Saint Martin has 120, 60 Hz on the Dutch side, and 220 50Hz on the French side.  Go figure.



I’m going to need to double check now, but the main 50v is feed from an unregulated “+53v” on the first LV board, and that was flat. Now I’m doubting what I saw.
I'd be looking at that +53 volts, shouldn't ever be flat, should be bumpy at 100 Hz.  It's the lower parts that may be of interest.

I’ll need to get a big capacitor as Mark suggested as first.
I’d post photos of the waveforms if I know how to add them here.
Oh, no attachments on the list, so you'd need to go to the website, and create a photo album in the photo section, then upload your files to there.

Harvey






Harvey White
 

Ah, I bet if you looked at waveform #1 and waveform #2 at the same time, the large dip in the waveform would match the smaller dip.

The amount of ripple on waveform #1 is excessive.  I'd suspect one of the rectifiers (since this is going to be a full wave, but not bridge (even though the part may be a bridge).  The other option is that the main filter capacitor is rather open.  Thinking about it a bit, I'd suspect the filter, since the waveform wouldn't be at 100 Hz without looking at both halves of the transformer sine wave.

Harvey

On 1/11/2022 3:56 PM, David Templeton wrote:
Ok, photos here of the misbehaving supplies: https://groups.io/g/TekScopes/album?id=271615

+130v and +40v measured on the rectifier board A11
And +50v measured on the regulator board A12