5000uf is not a standard capacity. But couldn't you just get the next up in value?
http://www.mouser.com/ProductDetail/Nichicon/LLS1E562MELZ/?qs=sGAEpiMZZMsh%252b1woXyUXj%252bATRo13k2ktCQydNqPeUTQ%3d
I looked at that cap, but it doesn't seem to meet the tolerance specs of the original. It's a 20% tolerance cap and at 5600UF, it could fall below the Tek spec of 5000UF +100 -0%. So, I didn't think that was a good choice. The way I read the specs, a cap that stays in the 5000UF to 10,000UF range is adequate, but I'm not sure what the impact would be to replace the cap with a 6000UF with 10% tolerance or 7500UF cap with 20% tolerance.
I looked at that cap, but it doesn't seem to meet the tolerance specs of the original. It's a 20% tolerance cap and at 5600UF, it could fall below the Tek spec of 5000UF +100 -0%. So, I didn't think that was a good choice. The way I read the specs, a cap that stays in the 5000UF to 10,000UF range is adequate, but I'm not sure what the impact would be to replace the cap with a 6000UF with 10% tolerance or 7500UF cap with 20% tolerance.
Based on a previous reply on the Tekscopes group, 5600 or 6800 uF should be fine. The bridge must be blown too (assuming the fuse F4501 is okay) otherwise you would be getting unfiltered negative DC, not +0.5V.
I looked at that cap, but it doesn't seem to meet the tolerance specs of the original. It's a 20% tolerance cap and at 5600UF, it could fall below the Tek spec of 5000UF +100 -0%. So, I didn't think that was a good choice. The way I read the specs, a cap that stays in the 5000UF to 10,000UF range is adequate, but I'm not sure what the impact would be to replace the cap with a 6000UF with 10% tolerance or 7500UF cap with 20% tolerance.
Based on a previous reply on the Tekscopes group, 5600 or 6800 uF should be fine. The bridge must be blown too (assuming the fuse F4501 is okay) otherwise you would be getting unfiltered negative DC, not +0.5V.
Yes, fuse F4501 looks and tests good. The bridge rectifier is available and a low cost part, so I ordered several, because this part is used on two other rails as well.
Yes, fuse F4501 looks and tests good. The bridge rectifier is available and a low cost part, so I ordered several, because this part is used on two other rails as well.
I don't know why they put a 5A fuse in a circuit with a BR that's designed for 2.5A...
From experience, the bridge rectifier is the fuse...
Low tolerance value for the 5000uf 10% part is 4500uf. And for the 5600uf 20% is 4480uf. Don't think it's a big deal.
oops I just noticed the -0% part. my bad. You could measure it with an ESR meter, it will probably be above tolerances. Nichicon caps are pretty good. Or get the higher value still.
6800uf at 20% has a low tolerance value of 5440.
http://www.mouser.com/Nichicon/Passive-Components/Capacitors/Aluminum-Electrolytic-Capacitors/_/N-75hqt?P=1z0wqtnZ1z0zlbmZ1yx4awy
There are a number of choices on that Mouser Electronics link. If I restrict the temp rating to 105C, the remaining prominent parameter is ripple current. What's the recommendation for optimum ripple current for a power supply filter cap?
I would suspect low ESR is an important parameter as well. I'll have to look at the individual data sheets to determine the choices with the best ESR rating.
Trying to select replacement caps for a recap is the tedious part of this project.
From experience, the bridge rectifier is the fuse...
That's the way I look at it. So, I bought six replacements, because I might blow another if there's a problem on one of the other boards.
An interesting observation while testing these filter caps in-circuit with my ESR meter was that they appeared to be two caps in series. The large filter caps appear to have three leads from the backside of the board. The three leads were aligned in a row, so I assumed the two outer leads were the negative leads. When I tested the caps between the positive and one of the negative leads, I got different measurements for capacitance between the two different negative leads. I know that capacitance tests aren't reliable in-circuit, but I didn't expect to see such a difference. I think the ESR varied slightly as well.
Low tolerance value for the 5000uf 10% part is 4500uf. And for the 5600uf 20% is 4480uf. Don't think it's a big deal.
oops I just noticed the -0% part. my bad. You could measure it with an ESR meter, it will probably be above tolerances. Nichicon caps are pretty good. Or get the higher value still.
6800uf at 20% has a low tolerance value of 5440.
http://www.mouser.com/Nichicon/Passive-Components/Capacitors/Aluminum-Electrolytic-Capacitors/_/N-75hqt?P=1z0wqtnZ1z0zlbmZ1yx4awy
There are a number of choices on that Mouser Electronics link. If I restrict the temp rating to 105C, the remaining prominent parameter is ripple current. What's the recommendation for optimum ripple current for a power supply filter cap?
I would suspect low ESR is an important parameter as well. I'll have to look at the individual data sheets to determine the choices with the best ESR rating.
Trying to select replacement caps for a recap is the tedious part of this project.
I linked multiple so that you can pick. 465B is from the time when low ESR didn't really exist. So I doubt it would make much difference.
My understanding is, that low ESR really helps when you're trying to filter high frequency switching rails. Switching regulators in recent years have really upped the frequency because this allows for smaller components to be used, but with this high frequency 200Khz+ you need caps with lower ESR to filter the ripple.
Lower ESR caps do tend to be rated for longer life as well.. so it might be worth it for that. Same is true of 105C rated caps.
For this replacement I would probably just go with a standard Nichicon (there are other good brands but I trust Nichicon the most) 85C cap.
I noted that the ESR measurements for the existing caps that could be measured had low ESR values. I know that I'm not dealing with a switching or high freq type power supply, but if the original caps appear to have low ESR then I would think the best choice replacement would be similar.
I've had good luck with Nichicon. I like the Rubycon as well.
I think that if there's a choice between a 85C cap and a 105C cap, the 105C is the better choice.
What's also interesting is that the two filter caps, C4429 & C4439, used for the 110V/55V bridge rectifier, CR4531, are rated at 100V even though there appears to be a test voltage at their location at 150.6V. It was my understanding that the caps need to have a voltage rating at or above anything they might experience, because that cap voltage rating is the max allowable.
Are you sure they haven't already been replaced (incorrectly)? There were solid tantalum axial caps in mine across the rails which were rated correctly but had exploded. Some chump may have stuck some underrated ones in.
I got those cap specs from the Tek Instruction Manual. These filter caps are radial.
What's also interesting is that the two filter caps, C4429 & C4439, used for the 110V/55V bridge rectifier, CR4531, are rated at 100V even though there appears to be a test voltage at their location at 150.6V. It was my understanding that the caps need to have a voltage rating at or above anything they might experience, because that cap voltage rating is the max allowable.
The capacitors are in series with +70.5V measured at the mid-point so +80.1V across the other. Interesting if one of them goes short though.
What's also interesting is that the two filter caps, C4429 & C4439, used for the 110V/55V bridge rectifier, CR4531, are rated at 100V even though there appears to be a test voltage at their location at 150.6V. It was my understanding that the caps need to have a voltage rating at or above anything they might experience, because that cap voltage rating is the max allowable.
The capacitors are in series with +70.5V measured at the mid-point so +80.1V across the other. Interesting if one of them goes short though.
I was thinking about a shorted cap as well. But, even without the short, C4439, rated at 100V, appears to see 150.6V on the positive side.
I was thinking about a shorted cap as well. But, even without the short, C4439, rated at 100V, appears to see 150.6V on the positive side.
Referenced to +70.5V, not ground, so only 80.1V across it. If C4429 goes short, then yes it will have the entire 150.6V across it.
I happened to come across a little piece of gear that claims to be most of what is needed to calibrate one of these tek scopes. I have no experience with it, but it may be enough to get a scope cal'd enough for moderate hobbyist use. If anyone on here has any experience with this thing, it would be great to know if it's worth dropping a little over 200USD on... But since I know you were looking for an "all in one" style calibrator, I thought I'd at least let you know about it.
http://www.ebay.com/itm/CALIBRATOR-AND-TESTER-FOR-TEKTRONIX-OSCILLOSCOPES-NEW-WARRANTED-CALIBRATED-/131638462260?hash=item1ea6438734:g:ZqQAAOSw14xWLxEy
That is an interesting device. It would be good to know what it can and can't do per the full Tek calibration procedure and if it meets all the Tek specs. That'll require some research. One good thing about it is that it's not made in China.
I was thinking about a shorted cap as well. But, even without the short, C4439, rated at 100V, appears to see 150.6V on the positive side.
Referenced to +70.5V, not ground, so only 80.1V across it. If C4429 goes short, then yes it will have the entire 150.6V across it.
So, if C4429 goes short, you'll lose both caps and likely make a mess.
I think if any of those two caps go short, then you've got 150V across the other.
I happened to come across a little piece of gear that claims to be most of what is needed to calibrate one of these tek scopes. I have no experience with it, but it may be enough to get a scope cal'd enough for moderate hobbyist use. If anyone on here has any experience with this thing, it would be great to know if it's worth dropping a little over 200USD on... But since I know you were looking for an "all in one" style calibrator, I thought I'd at least let you know about it.
http://www.ebay.com/itm/CALIBRATOR-AND-TESTER-FOR-TEKTRONIX-OSCILLOSCOPES-NEW-WARRANTED-CALIBRATED-/131638462260?hash=item1ea6438734:g:ZqQAAOSw14xWLxEy
That is an interesting device. It would be good to know what it can and can't do per the full Tek calibration procedure and if it meets all the Tek specs. That'll require some research. One good thing about it is that it's not made in China.
From the looks of it, if you wanted to follow the exact manual calibration procedure, it won't do anything that you want it to do. From what I understand about it from the description, controls, and output, it's more of a non-storage, CRT scope checker that you can use to make some vertical and horizontal adjustments from. However it does not look like it's capable of putting out the fast rise signal needed to adjust the HF comp of the verticals and trigger. It does have the advantage of being freshly built, calibrated, and tested - unlike the vintage cal gear you would need to purchase, repair, calibrate, and test in order to follow the manual spec procedure.
"The PG can be used to reliabily set vertical amplitude calibration, horizontal time calibration and to test triggering
capabilities at each of the ten steps up and the 10 steps down when the MODE switch is in the STEP position. "
(From the description)
If you enjoy repairing this scope, then by all means - buy and restore some old Tek cal equipment! That is what I am doing as my time and funds permit. If you want to take chances and buy "as is" or "power tested" cal gear, you can expect to pay $75 to $200 per piece of gear.. if you want it repaired and cal'd you're going to spend a small fortune. You've quite the decision to make in that aspect.
It's likely just good for making adjustments not calibration to Tek specs.
Hi
Back in the 1960's a lot of the gear you needed to calibrate things was pretty exotic stuff. Today fast rise time edges are a pretty normal thing to run into. A conventional (pick any brand) very high speed CMOS gate will give you a fast edge and a peak to peak that directly relates to the Vcc on the logic. Put a DVM on the supply and you know the cal signal you are getting.
Timing wise, a cheap packaged crystal oscillator will be much more accurate than the scope. Dividing that down to useful periods is a "play for the afternoon" sort of project on a cheap MCU board.
That certainly does not eliminate a high voltage DVM from your list of needed tools. It also does not take care of everything and anything you ever will need to do a full cal. It does take care of a lot of it though.
Bob