开云体育

I don't put the lid back on.
For me, this only serves to avoid purchasing adapters.
The use of an adapter when repairing a tektronix oscilloscopes enables a good and reliable repair.
It is possible to install modern electrolytes without an adapter,
you just have to pay attention to the fact that the body of the can serves as a jumper
for several points of ground potential. In that case you need to install the wires.
My opinion is that it might be better not to put the can back,
because the electrolyte has better cooling.
I don't know if I'm right with this interpretation, when it comes to the temperature of the electrolyte.
I know that restoring the can is more often used by restorers of old radios,
because in that case the original appearance is still more important.

"is NOS ok or brand spanking new"

Great question!

There are good reasons why a cap is spec'd for 85°C or 105°C, or 1000hrs or 5000hrs, or a shelf life of 10,000hrs or 100,000hrs.
It goes well beyond the electrolyte formula.
There's also the quality of the rubber seals, and the actual foil composition.

Simply put, a low-spec NOS 85°C cap is one you don't want to install after about 3-5 years on the shelf.
Whereas a high-spec NOS 105°C or 125°C is generally good for about 10+ years on the shelf.

Regarding brand-new - if it's a well-established series, then certainly preferable.
But if it's a recently introduced series, I normally wait about 2-3 years before using them.

I remember that I had serious problems with Panasonic FM's failing in the early years.
That really surprised me, being a lifelong fan of Matsushita/Panasonic.
No bulging, just capacitance decreasing and high ESR.
No, they weren't fakes.
I had a lot of comebacks, so I stopped using them for a couple of years.
Later, I returned to using them with no problems.

And Chemicon also had similar problems with a high-spec 105°C KYx or KZx series - I don't remember which.

So, crap happens even at the best people, and problems need time to be ironed out.


Beyond that, on PSU power rails, you GENERALLY do NOT have to replace an old cap with the same numbers.
For example, an old large 47uF can generally be replaced with a small 100uF or 220uF with no issues.
(Perhaps that's why <100uF caps are becoming scarce now).
The cap's voltage rating can be 2-3 times the actual circuit voltage with no issues.
Of course, you should check and understand the actual circuit requirements, before making any decision to change the original ratings.

Menahem Yachad
www.condoraudio.com

Well, it looks like I can indeed use a regular PT100 temp probe with this "option 2" instrument. If I put 110 ohms across the LEMO connector, I get a reading of 25 celsius, which aligns with the PT100 chart.

I remember getting the EOL notices a couple years ago. I guess it's finally here. FWIW, they still have a few thousand in most of those UHE and values at Mouser. I still have plenty Nichicon in the bins so we'll see how long that lasts me.

I spend about 20% of the repair characterizing and sorting components before I start so I understand the predicament. I've been getting manufacturer sample kits to see if there's anything that impresses me. Here's what I've been preparing to transition to:

Chemi-Con KYB Series:
- The Good - All around excellent, Low Z, Long life, Lots of sub 100?F values, High ripple, good prices
- The Bad - Once you get to 100V they become pricey
Chemi-Con PSG: (Organic Polymer Caps)
- The Good - Super low Z, Extremely long life, Similar price range, Excellent ESR!
- The Bad - Kind of wonky values to what we're used to. Not much below 100?F. No high voltage.
Chemi-Con KZN:
- The Good - Low Z, Long life, Excellent ESR and high ripple, Many sub 100?F values
- The Bad - Slightly pricier, Sub 100?F are ≤5000 hours rated,
Panasonic FR & FM Series:
- The Good - Low Z, Excellent prices, all 105°C, lots of sub 100?F options, Super low ESR
- The Bad - Not much really. I think they're ugly, does that count?
Cornell Dublier KXM & KBM Series:
- The Good - Hits all the marks, low Z, low ESR, good price, lots of sub 100?F values,
- The Bad - Also ugly, maybe uglier than the Panasonic. Otherwise, a fine cap.
Barker-Microfarad
- The Good - Everything, especially the 510D and made in USA! I thought my leakage tester was dead when I first sorted them.
- The Bad - Impossible to acquire by hobbyists like myself. A friend who works in component ordering for Mitsubishi got me a stash. Also, they don't look cheap. (From the style of them, they must have acquired the licensing from Sprague or are Sprague or manufacture the Sprague line for Vishay. I'm not really sure but you'll know what I mean when you see them.)

I'm sure there are many more. I might add I've been shying away from the Ruby's lately. They used to be another go to brand for me and many others. But during characterization, they tend to vary a lot more on my bench than the other brands here. It may be picky, but I don't like unpredictability.

Regards,
Drew

Some wise person said, "Why do we need faster and faster hardware?? Slower and slower software of course!? And it's even worse than that; software is getting slower and slower faster than hardware is getting faster and faster."? IEEE Spectrum magazine has an article in the April 2024 issue about lean and green software.? ?It's about time.? ? ? ? ? ? Jim

Yahoo Mail: Search, Organize, Conquer

On Sun, May 12, 2024 at 6:29 PM, Bruce Atwood<CCDman1@...> wrote: Everyone has to find there own solution but I have three identical laptops, all bought? used. (I bought the first one to receive a clone of the one I was using before I retired, they did not give it to me as a going away present.) One is XP with a 1TB HDD and about 70 Apps.? I would be hard pressed to re-install many of the Apps so I "backup the HDD by cloning it.? I also have a Windows 10 32 bit machine with a 512 GB SSD for routine email and browsing.? Again, I back it up by cloning it to another SSD.? Note that I have not been able to find a definite answer as to how long a SSD holds it data with no power so I try to clone it monthly.? And finally, since Turbo tax has screwed all of its long time users in a variety of ways, including insisting on a 64 bit OS, I now also have a 64 bit W10 machine.? Note that all three laptops are? Dell E4310s and while not exactly identical the HDDs or SDDs will boot fine in any of them thus backing me up for a laptop failure.? Hope I haven't jinks myself by writing down all this.? I haven't tried accessing the new and improved TekScopes from my XP machine but it seems to work fine from my W10/32 Bit with ThunderBird for mail and FireFox for browsing.

As to why all this is necessary there are some real improvements in security with the newest processors and auxiliary chips but I suspect that both the hardware and software firms motives are not pure as the driven snow.? Both of them make money selling replacements.? The software folks say you must have the latest OS for our latest and greatest software and the hardware guys can say you need new hardware to run the latest OS.? There ought to be a law.
cheers

Everyone has to find there own solution but I have three identical laptops, all bought used. (I bought the first one to receive a clone of the one I was using before I retired, they did not give it to me as a going away present.) One is XP with a 1TB HDD and about 70 Apps. I would be hard pressed to re-install many of the Apps so I "backup the HDD by cloning it. I also have a Windows 10 32 bit machine with a 512 GB SSD for routine email and browsing. Again, I back it up by cloning it to another SSD. Note that I have not been able to find a definite answer as to how long a SSD holds it data with no power so I try to clone it monthly. And finally, since Turbo tax has screwed all of its long time users in a variety of ways, including insisting on a 64 bit OS, I now also have a 64 bit W10 machine. Note that all three laptops are Dell E4310s and while not exactly identical the HDDs or SDDs will boot fine in any of them thus backing me up for a laptop failure. Hope I haven't jinks myself by writing down all this. I haven't tried accessing the new and improved TekScopes from my XP machine but it seems to work fine from my W10/32 Bit with ThunderBird for mail and FireFox for browsing.

As to why all this is necessary there are some real improvements in security with the newest processors and auxiliary chips but I suspect that both the hardware and software firms motives are not pure as the driven snow. Both of them make money selling replacements. The software folks say you must have the latest OS for our latest and greatest software and the hardware guys can say you need new hardware to run the latest OS. There ought to be a law.
cheers

Lowered price by $60. 开云体育 members get another $10 off. Also the FG501 that was going to be used for testing purposes is now unavailable. Instead, a BK4040 will be used.

On Sun, 12 May 2024 at 18:00, Greg Muir via groups.io <big_sky_explorer=
[email protected]> wrote:

But this doesn’t address the elephant in the room… having to either
upgrade (sometimes purchase) legacy software or purchase new at a cost.
That is the big chunk of money.

Why would buying a new computer mean you must throw out the existing one?
Can't you run run old stuff on the existing computer, and new stuff on the
new computer?

Theree was a lot of legacy computers at my last real job. A IMB five slot
PC that ran DOS and contained the interface for our oldest IC programmer.
We had several HP computers with tape memory cartridges for production
testing of our oldest product line. along with a rack of test equipment
with a computer running Win 2.0. Our IT department informed me that they
were going to 'replace all that junk' . I informed them that anything that
wasn't networked was nut theirs to screw with. They slipped in one
Saturday, and 'upgraded' that rack mount system to the then current Win 95
without backing up anything. The custom software belonged to Scientific
Atlanta, but they had no backup and it only ran under Win 2.0.
We had no backup as well. I had found hundreds of mostly bad 3.5" floppies
that took over a week to find the required Win 2.0 files, then reconstruct
the test software. There were about 20 failed attempts to back it up from
years before, before they gave up and forgot about it. We couldn't ship any
of that SA Telemetry receiver without the printout from t6he SATE system,
per our contract.
In the mid '80sI spent a week rebuilding the control files for an
industrial computer used for a Cable TV company. It stored their program
guide and five other NTSC video channels on a single 8" floppy. The cheap
SOB system manager refused to spend the $8 a month for a new floppy. so by
the time it failed, it ripped the heads out of the primary disk drive. The
second drive had dutifully backed up a defective disk, so it cost us over
$2000 to do a repair, then a week of examining old disks onee sector at a
time to rebuild the command structure. The OEM was out of business, and
SMS, who built the external floppy system, refused to repair it. WE nearly
lost a $60,000 computer system due to his stupidity.
For those who never worked with 8" drives, the motor ran constantly, and
many programmers didn't implement the head lift function, to improve
operating time so disks wore out in about 90 days so you rotated a new one
into the system, each month, after cleaning the heads.

My point is to understand your requirements and be ready for failure. I
oftem bught two or more of the same used model, and ussed them as older
systems died. I could then swap out parts to keep one working, to support
older software.

I buy used Dell computers for companies doing asset recovery for businesses
doing major upgrades. So fr, I've never received one with problems.

On Sun, May 12, 2024 at 1:00?PM Greg Muir via groups.io <big_sky_explorer=
[email protected]> wrote:

Thanks to all who suggested purchase of a used computer instead of a new
one. Been there, done that. Had good and bad experiences. Some people
don’t mention that there may be intermittent problems or other issues with
a unit. For new Dell’s offerings range around $500 to $700 for a new
machine. That doesn’t necessarily include some of the extra bells and
whistles (rapidly approaching $1k).

But this doesn’t address the elephant in the room… having to either
upgrade (sometimes purchase) legacy software or purchase new at a cost.
That is the big chunk of money.

I relation to this there is the client issue who may have equipment
operating in the field that is supported by legacy software that may now no
longer be available due to the manufacturer no longer in business. I can’t
tell them to purchase new equipment if they cannot afford it. I recently
inquired tech services at one company for a legacy software upgrade that
was no longer available to find out that one of the agents remember having
that upgrade stashed in his software collection he saved from the dumpster.
Problem solved when a 3-1/2” floppy arrived in the mail (had security stuff
embedded on the disk to prevent copying or downloads).

Some of the legacy software media I am exposed to include floppies, PCMCIA
cards, even Casio cards & ZIP media, and a few other very esoteric items.
Equipment interfaces frequently include RS232 requiring laptops with mating
ports (USB dongles don’t necessarily work). I have a small laptop running
WIN 3.1 to help one client who has some DOS/WIN control software. Tried
emulators to get that to operate under later WIN OSs with no success.

In all many of these systems have been operating for many decades with few
problems and people don’t really have an incentive to change until
something really catches fire.

(Speaking of real legacy, I might also add that the US Air Force finally
migrated away from using 8 inch floppies in their Minuteman missile launch
control facilities five or so years ago. Legacy? A bit.)

Greg

I have been rebuilding a SC504, all done except for the CRT shield for the face plate 337-2203-00 anybody know where one is?

Mark


--------------------------------------------.
/ Don't take life so seriously, __/
/ it's not permanent. /
\_________.---------------------'
_ |/
oo\'
(__)\ _
\ \ .' `.
\ \ / \
\ '" \
\ ( ) \
'-| )--| :. \
| | | | \ '.
c__; c__; '-..'>.__

Yeah, the last time I saw 8 inch floppies was in 1987 when I was writing software (the only time I've done that for pay) on an Apple ][ system.? They were out of date even then.? ? ? ? ? ? ? ? ? ? Supporting multiple obsolete media types, interfaces, etc. is a big reason why I don't strike out on my own.? My nightmare is getting stuck spending most of my time fighting those fires....? ? ? ? ? Jim Ford?

Yahoo Mail: Search, Organize, Conquer

On Sun, May 12, 2024 at 10:00 AM, Greg Muir via groups.io<big_sky_explorer@...> wrote: Thanks to all who suggested purchase of a used computer instead of a new one.? Been there, done that.? Had good and bad experiences.? Some people don’t mention that there may be intermittent problems or other issues with a unit.? For new Dell’s offerings range around $500 to $700 for a new machine.? That doesn’t necessarily include some of the extra bells and whistles (rapidly approaching $1k).

But this doesn’t address the elephant in the room…? having to either upgrade (sometimes purchase) legacy software or purchase new at a cost.? That is the big chunk of money.

I relation to this there is the client issue who may have equipment operating in the field that is supported by legacy software that may now no longer be available due to the manufacturer no longer in business.? I can’t tell them to purchase new equipment if they cannot afford it.? I recently inquired tech services at one company for a legacy software upgrade that was no longer available to find out that one of the agents remember having that upgrade stashed in his software collection he saved from the dumpster. Problem solved when a 3-1/2” floppy arrived in the mail (had security stuff embedded on the disk to prevent copying or downloads).

Some of the legacy software media I am exposed to include floppies, PCMCIA cards, even Casio cards & ZIP media, and a few other very esoteric items.? Equipment interfaces frequently include RS232 requiring laptops with mating ports (USB dongles don’t necessarily work).? I have a small laptop running WIN 3.1 to help one client who has some DOS/WIN control software.? Tried emulators to get that to operate under later WIN OSs with no success.

In all many of these systems have been operating for many decades with few problems and people don’t really have an incentive to change until something really catches fire.

(Speaking of real legacy, I might also add that the US Air Force finally migrated away from using 8 inch floppies in their Minuteman missile launch control facilities five or so years ago.? Legacy?? A bit.)

Greg

Hi, I have two DM502As that supposedly have option 2 installed as per the sticker on top.

The issue is that the circuits meant to be omitted in option 2 are not omitted, it seems. U1004, CR1101, R1103, and other surrounding components that should have been omitted are there!

So I am wondering, can I buy a P6601 and use it with my DM502A and not worry about my DM502As not supporting the temperature measure option?

Photos: /g/TekScopes/album?id=295034

Thanks

Benjamin

Hello,

Update. Spending quite some time on trying to figure out what is going on, I believe I see what is wrong, but not clear how to fix it. Not sure how to include figures so I added a photo album and link there (/g/TekScopes/album?id=295033 ). Hope it's possible to read anyways!

The overcurrent shutdown is caused by the horizontal deflection circuit. For some reason Q205 isn’t driven hard enough to either fully short or open the horizontal yoke causing – guessing – the efficiency/Q factor of the Yoke and C130 (0.3 uF cap) to go down drastically.

Fig 1. Schematic (/g/TekScopes/photo/295033/3781903?p=Created%2C%2C%2C20%2C2%2C0%2C0)

From time to time, the circuit works up to a supply voltage of 17 V on +25V and the horizontal drive circuit (most other parts disconnected) draws around 500 mA. Voltage on J305.1 (Yoke high side) is well above 200 V with only a single voltage spike at the ‘retrace’ – otherwise well grounded by Q205. Current in the coil runs at above +-2.5 A with a very good looking sweep (steep in retrace, straight in the horizontal sweep). Interestingly (?), the base drive of Q205 is about 90 degrees off from the control signal of U350.1.
Below is a figure showing the control signal from U350, base voltage of Q205, collector voltage of Q205/yoke high side J305.1 (same node) and current through yoke. Voltage of collector Q205 is shown as divided by 10 in order to fit the graph. +25V line was in this case fed 15 V and total current draw was 500 mA. This was the only time I managed to capture all voltages and yoke current at the same time.

Noticeable that it takes approximately half of the time that the control signal is high before base voltage of Q205 drops.

Fig 2 Startup (/g/TekScopes/photo/295033/3781904?p=Created%2C%2C%2C20%2C2%2C0%2C0)

U350 creates an approximately 100 kHz rectangular wave signal (U350.1 is a darlington open collector output, pulled up by 820 Ohms, R355, from +21 V).

When not working properly, the collector voltage of Q205 (same node as J305.1) indeed has a spike, but also shows big ripple when not expected to.
It seems like driving Q205 is crucial for correct operation. When powering the driver circuit (ramping +25V) and U350.1 starts generating rectangular pulses, Q205 collector voltage (J305.1) initially shows spikes, but with ringing.

At what seems like a transition from one mode to another when increasing +25V voltage, the collector voltage ringing disappears and the single voltage spike rises while overall current consumption goes down.
Before transition, voltages similar to above figure looks like in the following figure. +25V is here fed 10 V and current draw is 380 mA. Collector voltage/Yoke high side (J305.1) only reaches some 70 V, but also ‘rings’ when not expected to. Base voltage of Q205 here follows U350.1 very well.

Fig 3. Startup (/g/TekScopes/photo/295033/3781905?p=Created%2C%2C%2C20%2C2%2C0%2C0)

After transition (when fortunate...) voltages look like in the following figure. Current consumption is still 380 mA, but voltage has been increased to 12 V. Collector voltage/Yoke high side (J305.1) here spikes at around 230 V and is 0 otherwise. Base voltage of Q205 is here delayed by roughly 90 degrees and dips strongly when shutting Q205 off.
All figures use the same scales for comparison.

Fig 4. High efficiency (/g/TekScopes/photo/295033/3781906?p=Created%2C%2C%2C20%2C2%2C0%2C0)

For some reason the transition to high efficiency, without any ringing at Q205 collector and distinct drive of base for Q205, very often doesn’t occur and current consumption rises quickly above comfortable limits.

By removing CR156, DC supply of yoke drive is removed and then drive of base voltage for Q205 softly transitions from in-phase with U350.1 to being approximately 90 degrees delayed at nominal voltage of +25V. Negative drive is also very strong at -7 V at highest. Looking at the previous graphs, base voltage goes down to even -14 V.

The graphs below shows control voltage (U350.1) and Q205 Base voltage for +25V fed from 8 to 25 V.

Fig 5. Control signal and Q205 Base (/g/TekScopes/photo/295033/3781907?p=Created%2C%2C%2C20%2C2%2C0%2C0)

If anyone can help in explaining what is going on and why the circuit won't transition to high efficiency, it would be great. I have pulled out and checked several components, including the transformer T325 and also replaced Q330 (with Q10, same type), but see no part that is really broken.

What puzzles me is that driving Q205 seems to work fine when yoke is not fed any DC, but will very seldom transition to good drive when there is a current running through the yoke and Q205. Possibly some feedback via collector base capacitance?

Has anyone seen this behaviour before? Any suggestions?

Regards,
Staffan

I'm looking for new. To be a bit more specific, I'm looking for low impedance, low ESR, long life (>=5000 hours), and at least 105-degree radials to use on power rails.

Thanks,
Barry - N4BUQ

Thanks to all who suggested purchase of a used computer instead of a new one. Been there, done that. Had good and bad experiences. Some people don’t mention that there may be intermittent problems or other issues with a unit. For new Dell’s offerings range around $500 to $700 for a new machine. That doesn’t necessarily include some of the extra bells and whistles (rapidly approaching $1k).

But this doesn’t address the elephant in the room… having to either upgrade (sometimes purchase) legacy software or purchase new at a cost. That is the big chunk of money.

I relation to this there is the client issue who may have equipment operating in the field that is supported by legacy software that may now no longer be available due to the manufacturer no longer in business. I can’t tell them to purchase new equipment if they cannot afford it. I recently inquired tech services at one company for a legacy software upgrade that was no longer available to find out that one of the agents remember having that upgrade stashed in his software collection he saved from the dumpster. Problem solved when a 3-1/2” floppy arrived in the mail (had security stuff embedded on the disk to prevent copying or downloads).

Some of the legacy software media I am exposed to include floppies, PCMCIA cards, even Casio cards & ZIP media, and a few other very esoteric items. Equipment interfaces frequently include RS232 requiring laptops with mating ports (USB dongles don’t necessarily work). I have a small laptop running WIN 3.1 to help one client who has some DOS/WIN control software. Tried emulators to get that to operate under later WIN OSs with no success.

In all many of these systems have been operating for many decades with few problems and people don’t really have an incentive to change until something really catches fire.

(Speaking of real legacy, I might also add that the US Air Force finally migrated away from using 8 inch floppies in their Minuteman missile launch control facilities five or so years ago. Legacy? A bit.)

Greg

On 5/11/2024 11:17 PM, Benjamin Badrakh wrote:

So I used cyanocrylateto glue them together. I stopped the fogging effect by putting the knob above a fan to constantly move air. This allows the cyanocrylate vapors to move away from the knob and not precipitate on the knob.
It worked perfectly.

Very clever.

Very cool. How do you reattach the top of the can cleanly? Is there a trick or do you just fill up the empty space with eg. Silicone?

is NOS ok or brand spanking new


On Sun, May 12, 2024 at 9:22?AM n4buq via groups.io <n4buq=
[email protected]> wrote:

Nichicon has discontinued a large part of their leaded capacitors. Their
ULD, UHE, and others, are no longer available in small-value, radial
types. I'm looking for good replacements for blown tantalums for values
under 100uF, etc.

Anyone have suggestions as to what to look for now?

Thanks,
Barry - N4BUQ

Nichicon has discontinued a large part of their leaded capacitors. Their ULD, UHE, and others, are no longer available in small-value, radial types. I'm looking for good replacements for blown tantalums for values under 100uF, etc.

Anyone have suggestions as to what to look for now?

Thanks,
Barry - N4BUQ

Hello everyone,

I've been carrying out expert assessments in the various fields of EMC for 40 years, and in the early years I was able to use a MONROE 175 electrostatic voltmeter to measure how people became charged. In 95 to 98% of cases, they were negatively charged.

But if there's one subject we couldn't care less about, it's whether the discharge is positive or negative. It's only the value of the peak voltage at which the person is charged that counts and has a direct influence on the effect and pain felt. I read at one point that the current value of relative humidity measured was only 20%. With such a low value, it's totally normal to become charged as soon as the person makes the slightest movement.

F1EKU