Test Equipment Anonymous (TEA) group therapy thread

[cyclin_al]

British TV advertising can still turn up something that's surprisingly tasteful, subtle and can even be genuinely entertaining. Vis, one that that's been actively liked in this household recently:

Please laugh with me.  It took me a rather long moment to realize AA was actually Automobile Association.  For the first bit, I was trying to make the link between getting on the road and Alcoholics Anonymous   

Over here the associations tack on an extra letter, such as Canadian CAA or American AAA...

[cyclin_al]

Holy crow!  The TinySA and NanoVNA are on the move!  December to now to go from Brampton to Mississauga.  It's a start but I'm glad they're on the move.  If they get here early next week, I'll unbox everything together and maybe shoot a video.

Wow!  You could probably walk that distance in 7 hours.  7 weeks is a bit much.

I would share your suspicion that there was no stock in Brampton, and were waiting for it to come from somewhere else.

[Cerebus]

Please laugh with me.  It took me a rather long moment to realize AA was actually Automobile Association.  For the first bit, I was trying to make the link between getting on the road and Alcoholics Anonymous   

Over here the associations tack on an extra letter, such as Canadian CAA or American AAA...

Hey, they seem to be advocating letting Rover get behind the wheel, so...

[25 CPS]

Holy crow!  The TinySA and NanoVNA are on the move!  December to now to go from Brampton to Mississauga.  It's a start but I'm glad they're on the move.  If they get here early next week, I'll unbox everything together and maybe shoot a video.

Wow!  You could probably walk that distance in 7 hours.  7 weeks is a bit much.

I would share your suspicion that there was no stock in Brampton, and were waiting for it to come from somewhere else.

I think they got caught out for that.  They racked up a negative feedback from another unhappy buyer over excessive shipping time sometime in the last couple of days between when I checked the status last and now, so I think they’re over promising and under delivering (literally!) with the promise of local inventory.

[mnementh]

Yeah, but if they were waiting for it to come from elsewhere... the logical place would be Asia. If so, howinnahell did the vendor get it at all, when they are turning all mail deliveries away at the border right now...? 

I have OTC medication from Amazon US and we have important business mails that were mailed Express; both have been in limbo for over a month. 

mnem

[med6753]

Flipped one of my audio streamers (the one for external DAC) for reasonable money.

Also ordered a Scopetrex PCB kit to play Vectrex games on analog oscilloscopes (to have another compelling reason to get an analog scope ...)

in the German section of the bay, there is a Tek 2465A (NAWTS):   
https://www.ebay.de/itm/1-St-Tektronix-2465A-350MHz-Oscilloscope-4-Kanale-gebr-g-Zust-R-2-/133631481520

 

mnem


https://www.google.com/search?-b-d&q=enabler

There is also a 2445B offered by the same seller:

https://www.ebay.de/itm/1-St-Tektronix-2445B-200MHz-Oscilloscope-4-Kanale-gebr-g-Zust-R-2-/402638274932



Greetings,

If anyone is considering these 2 go with the 2465A. Stay away from the 2445B even though it's cheaper. Why? Well other than the greater B/W of the 2465A (350MHz) vs the 2445B (200MHz) there are design issues.

The 2465A will have it's cal data in a memory chip employing an external battery which is easy to change out with an external power source attached during the removal/install process. My 2430 has the same arrangement and I've done the battery change successfully. The 2445B will have a Dallas chip with an internal battery which is a real PITA to deal with and finding a replacement Dallas chip or an equivalent is a nightmare.

The 2465A will have a pin-thru-hole A5 processor board which has NO tants on it. The 2445B will have an SMT A5 processor board full of leak prone tants which will ruin the board. Stay away if it hasn't already been re-capped.

[Saskia]

I know that Neomys Sapiens has a Tec 460 or 465 waiting for me to pick it up.

I am still tempted, but horribly out of space.

[bd139]

If anyone is considering these 2 go with the 2465A. Stay away from the 2445B even though it's cheaper. Why? Well other than the greater B/W of the 2465A (350MHz) vs the 2445B (200MHz) there are design issues.

The 2465A will have it's cal data in a memory chip employing an external battery which is easy to change out with an external power source attached during the removal/install process. My 2430 has the same arrangement and I've done the battery change successfully. The 2445B will have a Dallas chip with an internal battery which is a real PITA to deal with and finding a replacement Dallas chip or an equivalent is a nightmare.

The 2465A will have a pin-thru-hole A5 processor board which has NO tants on it. The 2445B will have an SMT A5 processor board full of leak prone tants which will ruin the board. Stay away if it hasn't already been re-capped.

This is probably the best advice on the Internet these days. So much less hassle. Why I jumped on that 2445A a while back too 

[BU508A]

Stumbled over this, a HP 5005B signature multimeter.

https://www.ebay.com/itm/154276310308



That's a nice one.

But what is a signature multimeter doing/good for? 

[VK5RC]

HP in the 1980s and microprocessors were 1st introduced, brought in a problem tracing solution that involves putting the device into signature mode. Various test points would produce a known digital signal or signature, which when analysed by a 'signature analyzer' and compared against known good or fault codes in the manual would 'simplify' problem solving!
If you are fault tracing one of these vintage TEs they can be helpful - I am not sure how much a modern Salea type device can take it's place.
Rob

[med6753]

If anyone is considering these 2 go with the 2465A. Stay away from the 2445B even though it's cheaper. Why? Well other than the greater B/W of the 2465A (350MHz) vs the 2445B (200MHz) there are design issues.

The 2465A will have it's cal data in a memory chip employing an external battery which is easy to change out with an external power source attached during the removal/install process. My 2430 has the same arrangement and I've done the battery change successfully. The 2445B will have a Dallas chip with an internal battery which is a real PITA to deal with and finding a replacement Dallas chip or an equivalent is a nightmare.

The 2465A will have a pin-thru-hole A5 processor board which has NO tants on it. The 2445B will have an SMT A5 processor board full of leak prone tants which will ruin the board. Stay away if it hasn't already been re-capped.

This is probably the best advice on the Internet these days. So much less hassle. Why I jumped on that 2445A a while back too 

Or hold out for an older 2465/2445 (No suffix). No cal battery at all. Non-volatile EAROM stores the cal data. But supposedly the EAROM has a finite access life cycle which is probably why my 2465DMS takes checksum errors on boot up in the option board EAROM. Luckily you can bypass and boot successfully and everything appears OK but I suspect the ACV accuracy on the DMM is FUBAR.   

[mnementh]

Which is what I suspect may have happened with my ex-NASA 2465. Pic is from just before we left Tejas. If I'm lucky, it's just getting its brain scrambled during POST by a few bad caps in the PSU, but it would be like the dozenth 2465 I've recapped, and never saw THAT before. 

mnem
 

[BU508A]

That is looking pretty ugly! 
Hope, you can fix this one day. I assume, currently it is locked away in this storage of yours?

[URI]

Stumbled over this, a HP 5005B signature multimeter.

https://www.ebay.com/itm/154276310308



That's a nice one.

..but 200€ starting bid is a bit over the top..!
Despite the fact that it was meant as a versatile service tool incorporating signature analysis and DMM, one would rarely need it today/could barely use it today.

For signature analysis you need the "good" reference signatures, mentioned in the service manual. Try to get matching FW and service manual these days..      

I have a 5006A signature analyser and its fascinating to see it work. An yes, it works. And yes, I have HP TE that has a signature mode for debugging.

But out of the TE I've repaired I didn't have the luck to have the right signatures to the revision of the individual TE...   

[beanflying]

In spite of some far less than stellar packaging the Fluke arrived unharmed  I was dreading mega crustiness but it seems to have never had batteries fitted 

Now for the bad bit seems to be stuck regardless of press buttons reading 1999 with the decimal point changing with the range selected. To tired to read the manual tonight if anyone knows of a likely cause feel free 

Other than that fairly clean and tidy on the outside.

Ugh that may be toast. This is a well known symptom of the analogue IC being bust. The 5V rail runs comedically high voltage as it's actually unregulated and was made in days when the mains voltage was a lot lower so it can burn out the unobtainium ASICs. Check the power supply voltages first on mains power. Watch out because live mains runs the entire board so be damn careful. Then check the clock is running at C14. Should be a 220nF film cap which controls the integrator time constant. if the voltages are ok then clean the switches and check all the tants for shorts if they bothered to fit them in yours.

Really not a fan of the 8000A if I'm honest. It's either that or the display board burns up 

So a quick test of the rails shows them being pulled low at around +-12V (should be 15) and the 5V rail is 3.3 so Cap/Tant testing/replacement seems to be a reasonable start point. Switches seem to be clean and switching correctly (thanks @med  ).

I have way less than zero need for it regardless but I didn't have one 

[med6753]

Which is what I suspect may have happened with my ex-NASA 2465. Pic is from just before we left Tejas. If I'm lucky, it's just getting its brain scrambled during POST by a few bad caps in the PSU, but it would be like the dozenth 2465 I've recapped, and never saw THAT before. 

mnem
 

Well, if you ever do get it out of storage and determine the processor itself is hosed and it's not PSU related I do have a spare A5 board out of a 2465DMS which I think is good. (I can check it in one of my 2465's).

You can have it for cost of customs bullshit and shipping if it comes to that.

[med6753]

In spite of some far less than stellar packaging the Fluke arrived unharmed  I was dreading mega crustiness but it seems to have never had batteries fitted 

Now for the bad bit seems to be stuck regardless of press buttons reading 1999 with the decimal point changing with the range selected. To tired to read the manual tonight if anyone knows of a likely cause feel free 

Other than that fairly clean and tidy on the outside.

Ugh that may be toast. This is a well known symptom of the analogue IC being bust. The 5V rail runs comedically high voltage as it's actually unregulated and was made in days when the mains voltage was a lot lower so it can burn out the unobtainium ASICs. Check the power supply voltages first on mains power. Watch out because live mains runs the entire board so be damn careful. Then check the clock is running at C14. Should be a 220nF film cap which controls the integrator time constant. if the voltages are ok then clean the switches and check all the tants for shorts if they bothered to fit them in yours.

Really not a fan of the 8000A if I'm honest. It's either that or the display board burns up 

So a quick test of the rails shows them being pulled low at around +-12V (should be 15) and the 5V rail is 3.3 so Cap/Tant testing/replacement seems to be a reasonable start point. Switches seem to be clean and switching correctly (thanks @med  ).

I have way less than zero need for it regardless but I didn't have one 

I'm not surprised. I've had several Flukes of same vintage (8600A, 8800A) with shorted electrolytics in the PSU. I did re-cap my 8000A even though PSU was OK. But the 8000A does drift quite a bit because of PSU design. So once you get yours sorted out if you want my PSU modification let me know.   

[beanflying]

In spite of some far less than stellar packaging the Fluke arrived unharmed  I was dreading mega crustiness but it seems to have never had batteries fitted 

Now for the bad bit seems to be stuck regardless of press buttons reading 1999 with the decimal point changing with the range selected. To tired to read the manual tonight if anyone knows of a likely cause feel free 

Other than that fairly clean and tidy on the outside.

Ugh that may be toast. This is a well known symptom of the analogue IC being bust. The 5V rail runs comedically high voltage as it's actually unregulated and was made in days when the mains voltage was a lot lower so it can burn out the unobtainium ASICs. Check the power supply voltages first on mains power. Watch out because live mains runs the entire board so be damn careful. Then check the clock is running at C14. Should be a 220nF film cap which controls the integrator time constant. if the voltages are ok then clean the switches and check all the tants for shorts if they bothered to fit them in yours.

Really not a fan of the 8000A if I'm honest. It's either that or the display board burns up 

So a quick test of the rails shows them being pulled low at around +-12V (should be 15) and the 5V rail is 3.3 so Cap/Tant testing/replacement seems to be a reasonable start point. Switches seem to be clean and switching correctly (thanks @med  ).

I have way less than zero need for it regardless but I didn't have one 

I'm not surprised. I've had several Flukes of same vintage (8600A, 8800A) with shorted electrolytics in the PSU. I did re-cap my 8000A even though PSU was OK. But the 8000A does drift quite a bit because of PSU design. So once you get yours sorted out if you want my PSU modification let me know.

I am seriously considering putting Batteries in it as I have 4 spare D Cells. This maybe should sort out some of the drift issues? Quick look at the circuit of the power supply is it is maybe a bit too simple for its own good.

[med6753]

In spite of some far less than stellar packaging the Fluke arrived unharmed  I was dreading mega crustiness but it seems to have never had batteries fitted 

Now for the bad bit seems to be stuck regardless of press buttons reading 1999 with the decimal point changing with the range selected. To tired to read the manual tonight if anyone knows of a likely cause feel free 

Other than that fairly clean and tidy on the outside.

Ugh that may be toast. This is a well known symptom of the analogue IC being bust. The 5V rail runs comedically high voltage as it's actually unregulated and was made in days when the mains voltage was a lot lower so it can burn out the unobtainium ASICs. Check the power supply voltages first on mains power. Watch out because live mains runs the entire board so be damn careful. Then check the clock is running at C14. Should be a 220nF film cap which controls the integrator time constant. if the voltages are ok then clean the switches and check all the tants for shorts if they bothered to fit them in yours.

Really not a fan of the 8000A if I'm honest. It's either that or the display board burns up 

So a quick test of the rails shows them being pulled low at around +-12V (should be 15) and the 5V rail is 3.3 so Cap/Tant testing/replacement seems to be a reasonable start point. Switches seem to be clean and switching correctly (thanks @med  ).

I have way less than zero need for it regardless but I didn't have one 

I'm not surprised. I've had several Flukes of same vintage (8600A, 8800A) with shorted electrolytics in the PSU. I did re-cap my 8000A even though PSU was OK. But the 8000A does drift quite a bit because of PSU design. So once you get yours sorted out if you want my PSU modification let me know.

I am seriously considering putting Batteries in it as I have 4 spare D Cells. This maybe should sort out some of the drift issues? Quick look at the circuit of the power supply is it is maybe a bit too simple for its own good.

Correct. The crux of the problem is the +5V rail is unregulated. But in order to utilize something like a 7805 to get a stable +5V you have to install another small transformer because the original doesn't have the voltage headroom to allow the 7805 to regulate. I suppose you could design a fancy boost circuit to use the original transformer but I decided on adding another. It's tight but there was room.

I also modified the +/- 15V circuits to use 7815 and 7915 regulators and got rid of the zener/pass transistors. 

[mnementh]

Guess you can't cut it much closer than that!

mnem
I really need to run out and get a Lotto ticket right now...

[beanflying]

Time to add some burley to the waters and bait a Ganged Hook in Brumby's Backyard  eBay auction: #265006750399 Of course you NEED another Scope .....

[0culus]

If anyone is considering these 2 go with the 2465A. Stay away from the 2445B even though it's cheaper. Why? Well other than the greater B/W of the 2465A (350MHz) vs the 2445B (200MHz) there are design issues.

The 2465A will have it's cal data in a memory chip employing an external battery which is easy to change out with an external power source attached during the removal/install process. My 2430 has the same arrangement and I've done the battery change successfully. The 2445B will have a Dallas chip with an internal battery which is a real PITA to deal with and finding a replacement Dallas chip or an equivalent is a nightmare.

The 2465A will have a pin-thru-hole A5 processor board which has NO tants on it. The 2445B will have an SMT A5 processor board full of leak prone tants which will ruin the board. Stay away if it hasn't already been re-capped.

This is probably the best advice on the Internet these days. So much less hassle. Why I jumped on that 2445A a while back too 

Or hold out for an older 2465/2445 (No suffix). No cal battery at all. Non-volatile EAROM stores the cal data. But supposedly the EAROM has a finite access life cycle which is probably why my 2465DMS takes checksum errors on boot up in the option board EAROM. Luckily you can bypass and boot successfully and everything appears OK but I suspect the ACV accuracy on the DMM is FUBAR.

Personally, I'm over my 2465B. It's just not that great of an analog 'scope compared to predecessors. And all those repairability issues. I'd far rather have 7854. Those are at least mostly repairable and offer a lot of neat functionality (way more than the clunky measurements in the 2400 series)

In other news, I received a 7904 service manual. This should make my troubleshooting easier since it has the pull out schematics.

[tautech]

In spite of some far less than stellar packaging the Fluke arrived unharmed  I was dreading mega crustiness but it seems to have never had batteries fitted 

Now for the bad bit seems to be stuck regardless of press buttons reading 1999 with the decimal point changing with the range selected. To tired to read the manual tonight if anyone knows of a likely cause feel free 

Other than that fairly clean and tidy on the outside.

Ugh that may be toast. This is a well known symptom of the analogue IC being bust. The 5V rail runs comedically high voltage as it's actually unregulated and was made in days when the mains voltage was a lot lower so it can burn out the unobtainium ASICs. Check the power supply voltages first on mains power. Watch out because live mains runs the entire board so be damn careful. Then check the clock is running at C14. Should be a 220nF film cap which controls the integrator time constant. if the voltages are ok then clean the switches and check all the tants for shorts if they bothered to fit them in yours.

Really not a fan of the 8000A if I'm honest. It's either that or the display board burns up 

So a quick test of the rails shows them being pulled low at around +-12V (should be 15) and the 5V rail is 3.3 so Cap/Tant testing/replacement seems to be a reasonable start point. Switches seem to be clean and switching correctly (thanks @med  ).

I have way less than zero need for it regardless but I didn't have one 

I'm not surprised. I've had several Flukes of same vintage (8600A, 8800A) with shorted electrolytics in the PSU. I did re-cap my 8000A even though PSU was OK. But the 8000A does drift quite a bit because of PSU design. So once you get yours sorted out if you want my PSU modification let me know.

I am seriously considering putting Batteries in it as I have 4 spare D Cells. This maybe should sort out some of the drift issues? Quick look at the circuit of the power supply is it is maybe a bit too simple for its own good.

Correct. The crux of the problem is the +5V rail is unregulated. But in order to utilize something like a 7805 to get a stable +5V you have to install another small transformer because the original doesn't have the voltage headroom to allow the 7805 to regulate. I suppose you could design a fancy boost circuit to use the original transformer but I decided on adding another. It's tight but there was room.

I also modified the +/- 15V circuits to use 7815 and 7915 regulators and got rid of the zener/pass transistors.

Would an option be to replace the bridge with Shockley's to gain a bit more voltage headroom for a Vreg ?

[beanflying]

snippity

Correct. The crux of the problem is the +5V rail is unregulated. But in order to utilize something like a 7805 to get a stable +5V you have to install another small transformer because the original doesn't have the voltage headroom to allow the 7805 to regulate. I suppose you could design a fancy boost circuit to use the original transformer but I decided on adding another. It's tight but there was room.

I also modified the +/- 15V circuits to use 7815 and 7915 regulators and got rid of the zener/pass transistors.

Would an option be to replace the bridge with Shockley's to gain a bit more voltage headroom for a Vreg ?

Maybe and I have a stash of them here to play with. I don't have any Tants and I am fairly sure I would need a -15V reg (might have a variable one) so likely I will work out a pack of bits and order them and a few experimental bits as well.

[med6753]

snippity

Correct. The crux of the problem is the +5V rail is unregulated. But in order to utilize something like a 7805 to get a stable +5V you have to install another small transformer because the original doesn't have the voltage headroom to allow the 7805 to regulate. I suppose you could design a fancy boost circuit to use the original transformer but I decided on adding another. It's tight but there was room.

I also modified the +/- 15V circuits to use 7815 and 7915 regulators and got rid of the zener/pass transistors.

Would an option be to replace the bridge with Shockley's to gain a bit more voltage headroom for a Vreg ?

Maybe and I have a stash of them here to play with. I don't have any Tants and I am fairly sure I would need a -15V reg (might have a variable one) so likely I will work out a pack of bits and order them and a few experimental bits as well.

Here's the documented modification utilizing a 2nd transformer. The 7815/7915 do not require heat sinking as the current is low. The 7805 does have a small heat sink. The mains fuse will have to be upgraded because the load from the additional transformer will cause it to blow. This mod has been in place for almost 3 years and it works great.