Hi Donald,
In 2011 I described how I made wiper contacts from a USB-A socket for my SA. After 11 years those contacts are still going strong.
Ton PA0TBR
Another plastic-disk finger contact symptom and cause.? This is the disk on the back of the step attenuator.
?
SYMPTOM: Intermittent: LEVEL mode meter reads zero, RF still output.
CAUSE: A1A1S1 Meter Atten Sw (pins 5/6/16) contact broke loose.? See SS13.
ACTION: Reattach with JB Weld.? Also blob up the other two contacts that haven¡¯t broken yet.
NOTE: Do not need to remove A1A1.? Unmount pot using mini-ratchet and #1 Phillips bit.
NOTE: Inner contact kills RF and meter.? Outer contact goofs up meter range annunciators when meter is in LEVEL mode, probably lights 0-3 regardless of attenuator.
?
JFYI
Dave Wise
?
PS: Here are titles and URLs of some threads discussing this.
?
HP8640B Gear Repair -
/g/HP-Agilent-Keysight-equipment/topic/24788359
8565A in process -
/g/HP-Agilent-Keysight-equipment/topic/24800901
HP 8558 Missing contact fingers ?-
/g/HP-Agilent-Keysight-equipment/topic/24811643
Need Advice on Repairing 8640B Switch Contacts -
/g/HP-Agilent-Keysight-equipment/topic/24814127
HP 8640B range switch fingers and late manual -
/g/HP-Agilent-Keysight-equipment/topic/24815391
Wanting to buy rotor switch contact fingers that fall off in vintage HP Spectrum Analyzers ?-
/g/HP-Agilent-Keysight-equipment/topic/24815512
spectrum HP 8558b good but missing part ?-
/g/HP-Agilent-Keysight-equipment/topic/33132015
HP 8558B spectrum analyzer plug-in rotary switch -
/g/HP-Agilent-Keysight-equipment/topic/29340855
?
?
A very similar style of these ¡®finger¡¯ contact is found in the 8011/12/13 pulse generators which can be found quite cheaply¡. How do i know¡ just repaired one¡ works lovely.
On 28 May 2022, at 22:01, n4buq <n4buq@...> wrote:
?
Other HP switches use those wipers.? I think some of the slide switches on the 8640B use them but I think there are others too.
From:
"Donald Prins" <dyhprins@...>
To: "HP-Agilent-Keysight-equipment" <[email protected]>
Sent: Saturday, May 28, 2022 10:06:01 AM
Subject: Re: [HP-Agilent-Keysight-equipment] HP8640B no rf out
Hi Ellud,
Can you tell me where one can obtain those wipers? I have a HP8640B that is missing 3 in total on the FM/RANGE switches. Your help will be much appreciated.
Donald?
PE1AHJ
|
Re: anyone have broken HP 5342a counter GPIB/HPIB i/f or know of equivalents
? not sure about the dining room table unless you got a real good wife or sufficient funds to bribe her ? I think the 5342/43 came out in the late 1970s early 80s so that would put this into the 40+ year range of age , the main limits if it comes to fixing things is the availability of parts or suitable replacements , although TTL or CMOS parts are still manufactured things get harder with ECL logic circuits , processors or MMICs especially if they were in TO package From: [email protected] < [email protected]> On Behalf Of Dave McGuire via groups.io Sent: Wednesday, June 1, 2022 9:44 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] anyone have broken HP 5342a counter GPIB/HPIB i/f or know of equivalents On 6/1/22 22:41, Samudra wrote: I'm not sure what "ham friendly" means, On 5/25/22 17:31,
Samudra wrote:
Well,
<>?is your answer, IMHO.? Also, for
these old (failing) equipment that use TTL/CMOS chips, it's not worth
it to setup a whole test/service lab just to get things working. If it
doesn't work, and can't be reasonably fixed, out it goes - or provided
to someone who needs the spare parts, until all useful life is
scavenged. That actually does sound like what happens at hamfests and
trade between hams. Oh good grief, this counter isn't THAT old. People fix stuff like this all the time. It's easier with a well-equipped lab, but people fix these things on their dining room tables every day. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: anyone have broken HP 5342a counter GPIB/HPIB i/f or know of equivalents
I have three, ...no extender cards.? Just a lot of teeny-weeny
clips and probes.
'Like your comment on 'dining room tables'.
Sometimes you just need to sprawl out.
Don N5CID
======================================================
On 6/1/22 9:44 PM, Dave McGuire wrote:
toggle quoted message
Show quoted text
On 6/1/22 22:41, Samudra
wrote:
???????
I'm not sure what "ham friendly" means, On 5/25/22
17:31,
??????? Samudra wrote:
Well,
?is your
answer, IMHO.? Also, for these old (failing) equipment
that use TTL/CMOS chips, it's not worth it to setup a
whole test/service lab just to get things working. If it
doesn't work, and can't be reasonably fixed, out it goes
- or provided to someone who needs the spare parts,
until all useful life is scavenged. That actually does
sound like what happens at hamfests and trade between
hams.
? Oh good grief, this counter isn't THAT old.? People fix
stuff like this all the time.? It's easier with a
well-equipped lab, but people fix these things on their
dining room tables every day.
?????????? -Dave
|
Re: anyone have broken HP 5342a counter GPIB/HPIB i/f or know of equivalents
On 6/1/22 22:41, Samudra wrote: I'm not sure what "ham friendly" means, On 5/25/22 17:31,
Samudra wrote:
Well, <>?is your answer, IMHO.? Also, for these old (failing) equipment that use TTL/CMOS chips, it's not worth it to setup a whole test/service lab just to get things working. If it doesn't work, and can't be reasonably fixed, out it goes - or provided to someone who needs the spare parts, until all useful life is scavenged. That actually does sound like what happens at hamfests and trade between hams. Oh good grief, this counter isn't THAT old. People fix stuff like this all the time. It's easier with a well-equipped lab, but people fix these things on their dining room tables every day. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: anyone have broken HP 5342a counter GPIB/HPIB i/f or know of equivalents
On Wed, May 25, 2022 at 06:16 PM, Dave McGuire wrote:
I'm not sure what "ham friendly" means, On 5/25/22 17:31, Samudra wrote:
Well,??is your answer, IMHO.? Also, for these old (failing) equipment that use TTL/CMOS chips, it's not worth it to setup a whole test/service lab just to get things working. If it doesn't work, and can't be reasonably fixed, out it goes - or provided to someone who needs the spare parts, until all useful life is scavenged. That actually does sound like what happens at hamfests and trade between hams.? 73 de N3RDX
|
Re: E4407B strange peak shapes at low RBW
ok i just read your post on this and you stated that your issue occured after a processor board repair.....
HP as mentioned before uses EEPROMs on each module to store calibration information , the EEPROM content is normally read out and then stored in battery backed RAM for daily use by the firmware , more than likely when you fixed the CPU board the memory content got corrupted and as a result your filter response and probably some other data that was not as obvious was off ,? when you performed the described software operation you cleared the memory and reloaded the calibration data from the boards which subsequently cleared up your problem !
What surprises me a little though is that the analyzer did not reload the good data automatically ,? most instruments i worked on stored a checksum of the memory contents and if the checksum did not line up they pulled the data stored in EEPROM, it could be that they did things differently on the ESA because the unit has multiple EEPROMS ....
|
Re: E4407B strange peak shapes at low RBW
I am simply stating that for my unit showing same symptoms this ¡°did clear the issue on my unit¡±. ? Sandra ?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of Lothar baier Sent: Wednesday, June 1, 2022 9:07 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] E4407B strange peak shapes at low RBW ? Usually the security sanitization requirement ONLY applies to user data which means every piece of data that was stored by the user it does NOT apply to factory data ! What this means in real life is that all memory that is accessible by the user to store memory has to be either cleared or sanitized , this therefore applies to battery backed RAM , flash as well as HDD. The calibration data for each module however is stored in EEPROM on each module making it possible to change a assembly without having to completely recalibrate the unit , since this is not considered user data no clearing or sanitizing is required ? ? Going from memory is not always the best thing,? it¡¯s the NISPOM security sanitization of the instrument. Its in this document E4401-90513 - Security Features I¡¯ll see if it attaches to my reply ? It does require you to reload all firmware so you have to have the full set of floppies after you do it. ? I assume you¡¯ve tried a align all already and it didn¡¯t help. I can not guarantee this is fix your issue,? only that it fixed mine. ? Sandra ? ? ? ? Hi Sandra,
What is a NSCOM reset?
Do you mean factory reset / preset on the machine itself or some command?
|
Might be a possibility , I have some silver filled solder paste that was intended to be used for die attach but I have to find the data sheet to see the thermal resistance
?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]>
On Behalf Of Michael Perkins via groups.io
Sent: Wednesday, June 1, 2022 7:45 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging
?
How about silver solder? Silver is twice as good as aluminum.
?
On Wed, Jun 1, 2022 at 4:36 PM Tom Lee <tomlee@...> wrote:
The numbers are even worse than I cited. I happened to remember the
ratio of thermal resistances for copper and lead was about an order of
magnitude, and assumed that lead and solder would have similar values. I
just looked up the numbers to see how far off my assumption was, and it
happens that solder's thermal resistance is 4x that of lead. So solder
is something like 50x worse than copper. Not even worth trying.
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 6/1/2022 16:13, Bruce wrote:
> Probably should have thought of that B4 hitting send :-)
>
> Any possibility of a top side (or less possibly bottom side) heat sink
>
> Have you done the thermal analysis and are sure that "fixing" the vias
> will solve the problem
>
> Interesting discussion - lots of stuff to learn here :-)
>
> Cheers!
>
> Bruce
>
> Quoting Lothar baier <Lothar@...>:
>
>> The problem with solder is that the thermal conductivity of solder sucks
>>
>> -----Original Message-----
>> From:
[email protected]
>> <[email protected]> On Behalf Of Bruce via
>>
>> Sent: Wednesday, June 1, 2022 3:11 PM
>> To: [email protected]
>> Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging
>>
>> Could you fill them with solder without causing problems ???
>>
>> Perhaps a low melting point solder ?
>>
>> Cheers!
>>
>> Bruce
>>
>> Quoting Lothar baier <Lothar@...>:
>>
>>> i am working on a circuit board design for a X-Band PA , the part is
>>> in a QFN package with a ground slug , i put vias in the ground slug as
>>> common for ground and thermal reasons and taped off the vias on the
>>> bottom side with capton tape to minimize? solder wicking , the
>>> problem however i run into is that the vias are not providing enough
>>> thermal conductivity to ensure the die not exceeding the max
>>> temperature ratings.
>>> the apps engineer states that they recommend copper filled vias or
>>> using a coin and on the next round i can surely have the vias plugged
>>> but i am looking for a solution to salvage the prototype boards and
>>> evaluate the impact of plugged vias .
>>> copper paste of course is available and used as a anti seize agent for
>>> car repair shops but the question is whether this kind of paste offers
>>> sufficient thermal properties and whether there are any concerns using
>>> this for a circuit board ?
>>>
>>>
>>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
>
>
>
>
>
|
Re: E4407B strange peak shapes at low RBW
Usually the security sanitization requirement ONLY applies to user data which means every piece of data that was stored by the user it does NOT apply to factory data !
What this means in real life is that all memory that is accessible by the user to store memory has to be either cleared or sanitized , this therefore applies to battery backed RAM , flash as well as HDD.
The calibration data for each module however is stored in EEPROM on each module making it possible to change a assembly without having to completely recalibrate the unit , since this is not considered user data no clearing or sanitizing
is required
?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]>
On Behalf Of Sandra Carroll via groups.io
Sent: Wednesday, June 1, 2022 7:52 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] E4407B strange peak shapes at low RBW
?
Going from memory is not always the best thing,? it¡¯s the NISPOM security sanitization of the instrument.
Its in this document E4401-90513 - Security Features
I¡¯ll see if it attaches to my reply
?
It does require you to reload all firmware so you have to have the full set of floppies after you do it.
?
I assume you¡¯ve tried a align all already and it didn¡¯t help.
I can not guarantee this is fix your issue,? only that it fixed mine.
?
Sandra
?
?
?
?
Hi Sandra,
What is a NSCOM reset?
Do you mean factory reset / preset on the machine itself or some command?
|
Re: E4407B strange peak shapes at low RBW
Going from memory is not always the best thing,? it¡¯s the NISPOM security sanitization of the instrument. Its in this document E4401-90513 - Security Features I¡¯ll see if it attaches to my reply ? It does require you to reload all firmware so you have to have the full set of floppies after you do it. ? I assume you¡¯ve tried a align all already and it didn¡¯t help.
I can not guarantee this is fix your issue,? only that it fixed mine. ? Sandra ? ? ?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of Kalle Kempe Sent: Wednesday, June 1, 2022 1:23 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] E4407B strange peak shapes at low RBW ? Hi Sandra,
What is a NSCOM reset?
Do you mean factory reset / preset on the machine itself or some command?
|
How about silver solder? Silver is twice as good as aluminum. Mike
On Wed, Jun 1, 2022 at 4:36 PM Tom Lee < tomlee@...> wrote: The numbers are even worse than I cited. I happened to remember the
ratio of thermal resistances for copper and lead was about an order of
magnitude, and assumed that lead and solder would have similar values. I
just looked up the numbers to see how far off my assumption was, and it
happens that solder's thermal resistance is 4x that of lead. So solder
is something like 50x worse than copper. Not even worth trying.
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 6/1/2022 16:13, Bruce wrote:
> Probably should have thought of that B4 hitting send :-)
>
> Any possibility of a top side (or less possibly bottom side) heat sink
>
> Have you done the thermal analysis and are sure that "fixing" the vias
> will solve the problem
>
> Interesting discussion - lots of stuff to learn here :-)
>
> Cheers!
>
> Bruce
>
> Quoting Lothar baier <Lothar@...>:
>
>> The problem with solder is that the thermal conductivity of solder sucks
>>
>> -----Original Message-----
>> From: [email protected]
>> <[email protected]> On Behalf Of Bruce via
>>
>> Sent: Wednesday, June 1, 2022 3:11 PM
>> To: [email protected]
>> Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging
>>
>> Could you fill them with solder without causing problems ???
>>
>> Perhaps a low melting point solder ?
>>
>> Cheers!
>>
>> Bruce
>>
>> Quoting Lothar baier <Lothar@...>:
>>
>>> i am working on a circuit board design for a X-Band PA , the part is
>>> in a QFN package with a ground slug , i put vias in the ground slug as
>>> common for ground and thermal reasons and taped off the vias on the
>>> bottom side with capton tape to minimize? solder wicking , the
>>> problem however i run into is that the vias are not providing enough
>>> thermal conductivity to ensure the die not exceeding the max
>>> temperature ratings.
>>> the apps engineer states that they recommend copper filled vias or
>>> using a coin and on the next round i can surely have the vias plugged
>>> but i am looking for a solution to salvage the prototype boards and
>>> evaluate the impact of plugged vias .
>>> copper paste of course is available and used as a anti seize agent for
>>> car repair shops but the question is whether this kind of paste offers
>>> sufficient thermal properties and whether there are any concerns using
>>> this for a circuit board ?
>>>
>>>
>>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>
>
>
>
>
>
>
|
The numbers are even worse than I cited. I happened to remember the ratio of thermal resistances for copper and lead was about an order of magnitude, and assumed that lead and solder would have similar values. I just looked up the numbers to see how far off my assumption was, and it happens that solder's thermal resistance is 4x that of lead. So solder is something like 50x worse than copper. Not even worth trying.
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
toggle quoted message
Show quoted text
On 6/1/2022 16:13, Bruce wrote: Probably should have thought of that B4 hitting send :-)
Any possibility of a top side (or less possibly bottom side) heat sink
Have you done the thermal analysis and are sure that "fixing" the vias will solve the problem
Interesting discussion - lots of stuff to learn here :-)
Cheers!
Bruce
Quoting Lothar baier <Lothar@...>:
The problem with solder is that the thermal conductivity of solder sucks
-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Bruce via groups.io
Sent: Wednesday, June 1, 2022 3:11 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging
Could you fill them with solder without causing problems ???
Perhaps a low melting point solder ?
Cheers!
Bruce
Quoting Lothar baier <Lothar@...>:
i am working on a circuit board design for a X-Band PA , the part is
in a QFN package with a ground slug , i put vias in the ground slug as
common for ground and thermal reasons and taped off the vias on the
bottom side with capton tape to minimize? solder wicking , the
problem however i run into is that the vias are not providing enough
thermal conductivity to ensure the die not exceeding the max
temperature ratings.
the apps engineer states that they recommend copper filled vias or
using a coin and on the next round i can surely have the vias plugged
but i am looking for a solution to salvage the prototype boards and
evaluate the impact of plugged vias .
copper paste of course is available and used as a anti seize agent for
car repair shops but the question is whether this kind of paste offers
sufficient thermal properties and whether there are any concerns using
this for a circuit board ?
|
Re: HP 8510B Service manual CLIP
Then maybe your best choice is to try to purchase some spare boards - can you localize it to a board or two
Cheers!
Bruce
Quoting "tom_iphi via groups.io" <iphi@...>:
toggle quoted message
Show quoted text
I do have the 8510A schematics and I have tried to work with these. Power seems to be unchanged from 85101A to 85101B, but there seem to be a lot of changes in the digital cage.
Obviously, as already mentioned the bubble memory got replaced by EEPROM.
But as far as I could see, also the processor got an update.
And the IO card definitely uses different chips. So, no chance with the A schematics.
Best regards,
Tom DG8SAQ
|
Probably should have thought of that B4 hitting send :-)
Any possibility of a top side (or less possibly bottom side) heat sink
Have you done the thermal analysis and are sure that "fixing" the vias will solve the problem
Interesting discussion - lots of stuff to learn here :-)
Cheers!
Bruce
Quoting Lothar baier <Lothar@...>:
toggle quoted message
Show quoted text
The problem with solder is that the thermal conductivity of solder sucks
-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Bruce via groups.io
Sent: Wednesday, June 1, 2022 3:11 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging
Could you fill them with solder without causing problems ???
Perhaps a low melting point solder ?
Cheers!
Bruce
Quoting Lothar baier <Lothar@...>:
i am working on a circuit board design for a X-Band PA , the part is
in a QFN package with a ground slug , i put vias in the ground slug as
common for ground and thermal reasons and taped off the vias on the
bottom side with capton tape to minimize? solder wicking ,? the
problem however i run into is that the vias are not providing enough
thermal conductivity to ensure the die not exceeding the max
temperature ratings.
the apps engineer states that they recommend copper filled vias or
using a coin and on the next round i can surely have the vias plugged
but i am looking for a solution to salvage the prototype boards and
evaluate the impact of plugged vias .
copper paste of course is available and used as a anti seize agent for
car repair shops but the question is whether this kind of paste offers
sufficient thermal properties and whether there are any concerns using
this for a circuit board ?
|
We been down this road with parts at a former employer , people want more low cost packages that can be automatically placed with means QFN or Laminate packages but thermals are a concern with those especially since the power levels are increasing drastically .
Used to you had 1-2W amps in QFN packages now you have 1-25W parts , granted the efficiency is much better now than it used to be especially with GaN devices but you still got heat to remove .
Since this is not a mass product I could use a CP package part which bolts down on the heatsink but the CP packaged part runs typically $850 while the 5-20W parts in QFN/Laminate only run about $78-200 while offering more gain so there is a incentive to make it work
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] < [email protected]> On Behalf Of Tom Lee via groups.io Sent: Wednesday, June 1, 2022 4:45 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging It's incomplete to consider only the length of the thermal path. Lead has over an order of magnitude higher thermal resistance than copper, so its effect would be roughly equivalent to removing over 90% of the vias, and filling the remaining few with copper. --Tom -- Prof. Thomas H. Lee Allen Ctr., Rm. 205 350 Jane Stanford Way Stanford University Stanford, CA 94305-4070 On 6/1/2022 14:14, John Griessen wrote: On 6/1/22 14:41, Lothar baier wrote:
the thermal conductivity of solder sucks It's not a big thickness for it to go through though.? Maybe your heat
up problem is caused by the capton tape?
If you can solder to the under side, (maybe with a chisel shaped iron
in a metcal heater handle), add a piece of copper thin sheet in a L
shape, ( a fin), and it will flow heat away.
|
Gotcha !
What I was thinking is to screw down the board on the carrier plate/heat spreader and use a dispensing syringe to fill the holes with copper paste , then use a squeegee to remove the excess copper paste from the slug area , screen print
the solder paste and then solder the part down
?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]>
On Behalf Of Tom Lee via groups.io
Sent: Wednesday, June 1, 2022 4:41 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging
?
What I was referring to was the possibility of the paste ending up in places that you don't want and causing leakage/shorts.
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 6/1/2022 13:52, Lothar baier wrote:
Btw I don¡¯t think electrical conductivity is a primary concern as the plating of the vias is carrying the main currents , this is mainly about thermal conductivity
?
?
Well, you don't have too many options at this point, so give it a try. You aren't going to be shipping these, I assume.
The obvious problem to watch out for is the electrical conductivity of the paste. The thermal properties are likely to be roughly similar to those of the standard thermal pastes that are sold specifically for heatsinks. The results may be considered a worse-than-worst-case
test, so if the paste works, you can feel confident that copper-filled vias will work (and better).
In any case, good luck, Lothar!
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 6/1/2022 12:59, Lothar baier wrote:
i am working on a circuit board design for a X-Band PA , the part is in a QFN package with a ground slug , i put vias in the ground slug as common for ground and thermal reasons and taped off the vias on the bottom side with capton tape
to minimize? solder wicking ,? the problem however i run into is that the vias are not providing enough thermal conductivity to ensure the die not exceeding the max temperature ratings.?
the apps engineer states that they recommend copper filled vias or using a coin and on the next round i can surely have the vias plugged but i am looking for a solution to salvage the prototype boards and evaluate the impact of plugged vias .
copper paste of course is available and used as a anti seize agent for car repair shops but the question is whether this kind of paste offers sufficient thermal properties and whether there are any concerns using this for a circuit board ?
?
?
|
It's incomplete to consider only the length of the thermal path.
Lead has over an order of magnitude higher thermal resistance than copper, so its effect would be roughly equivalent to removing over 90% of the vias, and filling the remaining few with copper.
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
toggle quoted message
Show quoted text
On 6/1/2022 14:14, John Griessen wrote: On 6/1/22 14:41, Lothar baier wrote:
the thermal conductivity of solder sucks It's not a big thickness for it to go through though.? Maybe your heat up problem is caused by the capton tape?
If you can solder to the under side, (maybe with a chisel shaped iron in a metcal heater handle), add a piece of copper thin sheet in a L shape, ( a fin), and it will flow heat away.
|
What I was referring to was the possibility of the paste ending up
in places that you don't want and causing leakage/shorts.
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 6/1/2022 13:52, Lothar baier wrote:
toggle quoted message
Show quoted text
Btw I don¡¯t think electrical conductivity
is a primary concern as the plating of the vias is carrying
the main currents , this is mainly about thermal conductivity
?
?
Well, you don't have too many options at
this point, so give it a try. You aren't going to be shipping
these, I assume.
The obvious problem to watch out for is the electrical
conductivity of the paste. The thermal properties are likely
to be roughly similar to those of the standard thermal pastes
that are sold specifically for heatsinks. The results may be
considered a worse-than-worst-case test, so if the paste
works, you can feel confident that copper-filled vias will
work (and better).
In any case, good luck, Lothar!
--Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 6/1/2022 12:59, Lothar baier wrote:
i am working on a circuit board design
for a X-Band PA , the part is in a QFN package with a ground
slug , i put vias in the ground slug as common for ground
and thermal reasons and taped off the vias on the bottom
side with capton tape to minimize? solder wicking ,? the
problem however i run into is that the vias are not
providing enough thermal conductivity to ensure the die not
exceeding the max temperature ratings.?
the apps engineer states that they recommend copper filled
vias or using a coin and on the next round i can surely have
the vias plugged but i am looking for a solution to salvage
the prototype boards and evaluate the impact of plugged vias
.
copper paste of course is available and used as a anti seize
agent for car repair shops but the question is whether this
kind of paste offers sufficient thermal properties and
whether there are any concerns using this for a circuit
board ?
?
|
No I only put the capton tape on there for masking and removed it after the soldering was done , I also tried to improve the thermal interface between the board and the heat spreader including gap pads and copper but its not enough improvement .
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] < [email protected]> On Behalf Of John Griessen via groups.io Sent: Wednesday, June 1, 2022 4:14 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] OT: Via plugging On 6/1/22 14:41, Lothar baier wrote: the thermal conductivity of solder sucks It's not a big thickness for it to go through though. Maybe your heat up problem is caused by the capton tape? If you can solder to the under side, (maybe with a chisel shaped iron in a metcal heater handle), add a piece of copper thin sheet in a L shape, ( a fin), and it will flow heat away.
|
On 6/1/22 14:41, Lothar baier wrote: the thermal conductivity of solder sucks It's not a big thickness for it to go through though. Maybe your heat up problem is caused by the capton tape? If you can solder to the under side, (maybe with a chisel shaped iron in a metcal heater handle), add a piece of copper thin sheet in a L shape, ( a fin), and it will flow heat away.
|
Dave McGuire