|
I have E4435B signal gen from ESG series that loses output in certain parts of its output range 250..350MHz, 500..700MHz, etc.
The problem seems to be in the VCO or its PLL loop.?
Its output is supposed to be 500..1000MHz before further multiplication/division but lower part of the range is unstable. As you go down in frequency from 1000MHz to 500MHz, somewhere around 690MHz it becomes noisy and at 680MHz it loses signal altogether.
These siggens sometimes have prescaler U313 1GC1-4210 failure, but here it does not seem the case as it is further down the signal path - after YIG multiplier that converts 0.5-1GHz VCO output into 4-8GHz signal.
This instrument is a patchwork of modules and also has some calibration file integrity issues so I am trying to understand whether this is hardware or configuration problem. At the moment I am looking at low level operation of the PLL loop filter and its phase detector.? I only have block diagram to go with http://www.leobodnar.com/HP35650/E4435B%20A24%20block%20diagram.png
I have noticed that the synthesizer board A24 from ESG signal gen series is very similar to synthesizer board A19 from E4406A spectrum analyser.? Their VCO/PLL loops are almost identical.
A24 from E4435B: http://www.leobodnar.com/HP35650/E4435B%20A24%20SYNTHESIZER%20ASSEMBLY%20top.jpeg
A19 from E4406A: http://www.leobodnar.com/HP35650/E4406A%20A19%20SYNTHESIZER%20ASSEMBLY%20top.jpeg
Assuming that Agilent has been reusing their synthesizer design across many E4xxx class instruments, is there a detailed schematic or CLIP for the synthesizer module in other instruments of the same era?
Thanks
Leo
|
Hi Leo,
Which synthesizer board do you have?
There is a service note (E4436B-01F) that says: "Serial Numbers: [0000A00000 / 9999Z99999]" "The ESG-B AP/DP Family of Signal Generators has been experiencing failures in the Frac-N board assemblies. This is due to a failure in the divide-by-eight circuit of the Divider chip A24 U313 which causes the ESG to display Error 511 "Output unleveled". All ESG-B AP/DP units with the serial numbers mentioned above may experience Error 511 along with UNSTABLE or NO OUPUT below 1GHz. If this occurs, the A24 Frac-N/Divider Assembly (E4423- 60019) needs to be replaced. This assembly should be ordered using part number E4423-69019 and should be treated as a warranty repair. The defective assembly should then be returned with documentation referencing this service note as "Error 511"."
Tom Bryan
toggle quoted message
Show quoted text
On 11/11/2018 4:39 AM, Leo Bodnar wrote: I have E4435B signal gen from ESG series that loses output in certain parts of its output range 250..350MHz, 500..700MHz, etc.
The problem seems to be in the VCO or its PLL loop.
Its output is supposed to be 500..1000MHz before further multiplication/division but lower part of the range is unstable. As you go down in frequency from 1000MHz to 500MHz, somewhere around 690MHz it becomes noisy and at 680MHz it loses signal altogether.
These siggens sometimes have prescaler U313 1GC1-4210 failure, but here it does not seem the case as it is further down the signal path - after YIG multiplier that converts 0.5-1GHz VCO output into 4-8GHz signal.
|
Someone ask for ESG Fractional N divider. Not my and don't ask for better resolution.?
On Sun, 11 Nov 2018, 14:35 Tom B < tbryan@... wrote: Hi Leo,
Which synthesizer board do you have?
There is a service note (E4436B-01F) that says: "Serial Numbers:
[0000A00000 / 9999Z99999]" "The ESG-B AP/DP Family of Signal Generators
has been experiencing failures in the Frac-N board assemblies. This is
due to a failure in the divide-by-eight circuit of the Divider chip A24
U313 which causes the ESG to display Error 511 "Output unleveled". All
ESG-B AP/DP units with the serial numbers mentioned above may experience
Error 511 along with UNSTABLE or NO OUPUT below 1GHz. If this occurs,
the A24 Frac-N/Divider Assembly (E4423- 60019) needs to be replaced.
This assembly should be ordered using part number E4423-69019 and should
be treated as a warranty repair. The defective assembly should then be
returned with documentation referencing this service note as "Error 511"."
Tom Bryan
On 11/11/2018 4:39 AM, Leo Bodnar wrote:
> I have E4435B signal gen from ESG series that loses output in certain
> parts of its output range 250..350MHz, 500..700MHz, etc.
> The problem seems to be in the VCO or its PLL loop.
> Its output is supposed to be 500..1000MHz before further
> multiplication/division but lower part of the range is unstable. As
> you go down in frequency from 1000MHz to 500MHz, somewhere around
> 690MHz it becomes noisy and at 680MHz it loses signal altogether.
> These siggens sometimes have prescaler U313 1GC1-4210 failure, but
> here it does not seem the case as it is further down the signal path -
> after YIG multiplier that converts 0.5-1GHz VCO output into 4-8GHz signal.
>
>
|
So just out of curiosity where didi you find the schematics?? I thought there no schematics avaible anywhere for this device?
Il giorno mer 14 nov 2018 alle ore 23:41 Milan Vasic < vasa1958@...> ha scritto: Someone ask for ESG Fractional N divider. Not my and don't ask for better resolution.?
On Sun, 11 Nov 2018, 14:35 Tom B < tbryan@... wrote: Hi Leo,
Which synthesizer board do you have?
There is a service note (E4436B-01F) that says: "Serial Numbers:
[0000A00000 / 9999Z99999]" "The ESG-B AP/DP Family of Signal Generators
has been experiencing failures in the Frac-N board assemblies. This is
due to a failure in the divide-by-eight circuit of the Divider chip A24
U313 which causes the ESG to display Error 511 "Output unleveled". All
ESG-B AP/DP units with the serial numbers mentioned above may experience
Error 511 along with UNSTABLE or NO OUPUT below 1GHz. If this occurs,
the A24 Frac-N/Divider Assembly (E4423- 60019) needs to be replaced.
This assembly should be ordered using part number E4423-69019 and should
be treated as a warranty repair. The defective assembly should then be
returned with documentation referencing this service note as "Error 511"."
Tom Bryan
On 11/11/2018 4:39 AM, Leo Bodnar wrote:
> I have E4435B signal gen from ESG series that loses output in certain
> parts of its output range 250..350MHz, 500..700MHz, etc.
> The problem seems to be in the VCO or its PLL loop.
> Its output is supposed to be 500..1000MHz before further
> multiplication/division but lower part of the range is unstable. As
> you go down in frequency from 1000MHz to 500MHz, somewhere around
> 690MHz it becomes noisy and at 680MHz it loses signal altogether.
> These siggens sometimes have prescaler U313 1GC1-4210 failure, but
> here it does not seem the case as it is further down the signal path -
> after YIG multiplier that converts 0.5-1GHz VCO output into 4-8GHz signal.
>
>
-- Alberto Vaudagna IDRA-XAM Electronics Firmware and Software division manager- Team H2politO H2politO – Politecnico di Torino
|
My friend have in his company but internal security and copyright policy is very strict. No chance to make a copy of all schematics.?
So just out of curiosity where didi you find the schematics?? I thought there no schematics avaible anywhere for this device?
Il giorno mer 14 nov 2018 alle ore 23:41 Milan Vasic < vasa1958@...> ha scritto: Someone ask for ESG Fractional N divider. Not my and don't ask for better resolution.?
On Sun, 11 Nov 2018, 14:35 Tom B < tbryan@... wrote: Hi Leo,
Which synthesizer board do you have?
There is a service note (E4436B-01F) that says: "Serial Numbers:
[0000A00000 / 9999Z99999]" "The ESG-B AP/DP Family of Signal Generators
has been experiencing failures in the Frac-N board assemblies. This is
due to a failure in the divide-by-eight circuit of the Divider chip A24
U313 which causes the ESG to display Error 511 "Output unleveled". All
ESG-B AP/DP units with the serial numbers mentioned above may experience
Error 511 along with UNSTABLE or NO OUPUT below 1GHz. If this occurs,
the A24 Frac-N/Divider Assembly (E4423- 60019) needs to be replaced.
This assembly should be ordered using part number E4423-69019 and should
be treated as a warranty repair. The defective assembly should then be
returned with documentation referencing this service note as "Error 511"."
Tom Bryan
On 11/11/2018 4:39 AM, Leo Bodnar wrote:
> I have E4435B signal gen from ESG series that loses output in certain
> parts of its output range 250..350MHz, 500..700MHz, etc.
> The problem seems to be in the VCO or its PLL loop.
> Its output is supposed to be 500..1000MHz before further
> multiplication/division but lower part of the range is unstable. As
> you go down in frequency from 1000MHz to 500MHz, somewhere around
> 690MHz it becomes noisy and at 680MHz it loses signal altogether.
> These siggens sometimes have prescaler U313 1GC1-4210 failure, but
> here it does not seem the case as it is further down the signal path -
> after YIG multiplier that converts 0.5-1GHz VCO output into 4-8GHz signal.
>
>
--
Alberto Vaudagna IDRA-XAM Electronics Firmware and Software division manager- Team H2politO H2politO – Politecnico di Torino
|
Oh ok well he is lucky! I only see once the clip manual selling on eBay and I miss it. And of course I'm looking for that I guess we should ask to keysight ?
Il giorno gio 15 nov 2018, 12:03 Milan Vasic < vasa1958@...> ha scritto: My friend have in his company but internal security and copyright policy is very strict. No chance to make a copy of all schematics.?
So just out of curiosity where didi you find the schematics?? I thought there no schematics avaible anywhere for this device?
Il giorno mer 14 nov 2018 alle ore 23:41 Milan Vasic < vasa1958@...> ha scritto: Someone ask for ESG Fractional N divider. Not my and don't ask for better resolution.?
On Sun, 11 Nov 2018, 14:35 Tom B < tbryan@... wrote: Hi Leo,
Which synthesizer board do you have?
There is a service note (E4436B-01F) that says: "Serial Numbers:
[0000A00000 / 9999Z99999]" "The ESG-B AP/DP Family of Signal Generators
has been experiencing failures in the Frac-N board assemblies. This is
due to a failure in the divide-by-eight circuit of the Divider chip A24
U313 which causes the ESG to display Error 511 "Output unleveled". All
ESG-B AP/DP units with the serial numbers mentioned above may experience
Error 511 along with UNSTABLE or NO OUPUT below 1GHz. If this occurs,
the A24 Frac-N/Divider Assembly (E4423- 60019) needs to be replaced.
This assembly should be ordered using part number E4423-69019 and should
be treated as a warranty repair. The defective assembly should then be
returned with documentation referencing this service note as "Error 511"."
Tom Bryan
On 11/11/2018 4:39 AM, Leo Bodnar wrote:
> I have E4435B signal gen from ESG series that loses output in certain
> parts of its output range 250..350MHz, 500..700MHz, etc.
> The problem seems to be in the VCO or its PLL loop.
> Its output is supposed to be 500..1000MHz before further
> multiplication/division but lower part of the range is unstable. As
> you go down in frequency from 1000MHz to 500MHz, somewhere around
> 690MHz it becomes noisy and at 680MHz it loses signal altogether.
> These siggens sometimes have prescaler U313 1GC1-4210 failure, but
> here it does not seem the case as it is further down the signal path -
> after YIG multiplier that converts 0.5-1GHz VCO output into 4-8GHz signal.
>
>
--
Alberto Vaudagna IDRA-XAM Electronics Firmware and Software division manager- Team H2politO H2politO – Politecnico di Torino
|
Thanks to everybody for great discussion.? Despite initial diversion I have found that U313 Agilent 1GC1-4210 has struck yet again.? It had weird failure mode where it would change its division ratio at the bottom 30% of its input range of 4GHz..8GHz to half the intended value.? This caused its output to double in frequency and become filtered out by downstream LPFs.? So I chucked U313 out and replaced it with good honest organic grown 24GHz Hittite prescalers.? E4435B now works just how it was supposed to.? If there is enough interest I can create a drop-in PCB that will replace 1GC1-4210. Cheers Leo 
|
toggle quoted message
Show quoted text
On Fri, 16 Nov 2018, 21:28 Leo Bodnar < leo@... wrote: Thanks to everybody for great discussion.?
Despite initial diversion I have found that U313 Agilent 1GC1-4210 has struck yet again.? It had weird failure mode where it would change its division ratio at the bottom 30% of its input range of 4GHz..8GHz to half the intended value.? This caused its output to double in frequency and become filtered out by downstream LPFs.? So I chucked U313 out and replaced it with good honest organic grown 24GHz Hittite prescalers.? E4435B now works just how it was supposed to.? If there is enough interest I can create a drop-in PCB that will replace 1GC1-4210.
Cheers
Leo
|
I'd sign up for a replacement board or even just the gerbers so I could get my own made instead of using your time.
In fact I'd probably just be happy with the schematic, parts list and instructions if that was easier.
TonyG
|
Leo, Will be good when you have already open A24 module to write-down state of the pins 8, 9 and 10 of U313 in function of the frequency bands. Milan
I'd sign up for a replacement board or even just the gerbers so I could get my own made instead of using your time.
In fact I'd probably just be happy with the schematic, parts list and instructions if that was easier.
TonyG
|
well at the end I bought them, I offer 25 and I had 15 coupon so at the edn I spend 13 dollar in total! Thanks for the link!
Il giorno ven 16 nov 2018 alle ore 22:49 Milan Vasic < vasa1958@...> ha scritto: Leo, Will be good when you have already open A24 module to write-down state of the pins 8, 9 and 10 of U313 in function of the frequency bands. Milan I'd sign up for a replacement board or even just the gerbers so I could get my own made instead of using your time.
In fact I'd probably just be happy with the schematic, parts list and instructions if that was easier.
TonyG
-- Alberto Vaudagna IDRA-XAM Electronics Firmware and Software division manager- Team H2politO H2politO – Politecnico di Torino
|
Beautiful job!
toggle quoted message
Show quoted text
On 11/16/2018 4:32 PM, Milan Vasic wrote: Good job Leo!
Milan
On Fri, 16 Nov 2018, 21:28 Leo Bodnar <leo@... <mailto:leo@...> wrote:
Thanks to everybody for great discussion.
Despite initial diversion I have found that U313 Agilent 1GC1-4210 has
struck yet again.? It had weird failure mode where it would change its
division ratio at the bottom 30% of its input range of 4GHz..8GHz to half
the intended value. This caused its output to double in frequency and
become filtered out by downstream LPFs.? So I chucked U313 out and
replaced it with good honest organic grown 24GHz Hittite prescalers.?
E4435B now works just how it was supposed to.? If there is enough interest
I can create a drop-in PCB that will replace 1GC1-4210.
Cheers
Leo
|
Great job Leo
Certainly interest here for a drop in PCB, gerber files, the
copies of the schematics you worked out the existing functions or
even the chip number for the Hittite prescalar:- basically any
info available on what you did.
I have been thinking of doing the same for sometime but could
never find any schematics of the area around the prescaler and not
until Milan posted the area of this chip the other day I have
never seen anything on it.
Thanks
Dave
17/11/2018 4:28 AM, Leo Bodnar wrote:
toggle quoted message
Show quoted text
Thanks to everybody for great discussion.?
Despite initial diversion I have found that U313 Agilent 1GC1-4210
has struck yet again.? It had weird failure mode where it would
change its division ratio at the bottom 30% of its input range of
4GHz..8GHz to half the intended value.? This caused its output to
double in frequency and become filtered out by downstream LPFs.?
So I chucked U313 out and replaced it with good honest organic
grown 24GHz Hittite prescalers.? E4435B now works just how it was
supposed to.? If there is enough interest I can create a drop-in
PCB that will replace 1GC1-4210.
Cheers
Leo
|
I am making a trial run of few PCBs and I have a E4425B with "250MHz..350MHz unlock problem" to try it on. ? I will document the process as I do it. Cheers
On 20 Nov 2018, at 06:27, Dave Ireland wrote:
Great job Leo
Certainly interest here for a drop in PCB, gerber files, the
copies of the schematics you worked out the existing functions or
even the chip number for the Hittite prescalar:- basically any
info available on what you did.
I have been thinking of doing the same for sometime but could
never find any schematics of the area around the prescaler and not
until Milan posted the area of this chip the other day I have
never seen anything on it.
Thanks
Dave
17/11/2018 4:28 AM, Leo Bodnar wrote:
Thanks to everybody for great discussion.?
Despite initial diversion I have found that U313 Agilent 1GC1-4210
has struck yet again.? It had weird failure mode where it would
change its division ratio at the bottom 30% of its input range of
4GHz..8GHz to half the intended value.? This caused its output to
double in frequency and become filtered out by downstream LPFs.?
So I chucked U313 out and replaced it with good honest organic
grown 24GHz Hittite prescalers.? E4435B now works just how it was
supposed to.? If there is enough interest I can create a drop-in
PCB that will replace 1GC1-4210.
Cheers
Leo
<IMG_6824.jpg>
|
Update: I have made a successful prototype run of a 1GC1-4210 drop-in replacement PCB for the following instruments: 250 kHz – 1 GHz: E4423B, E4434B 250 kHz – 2 GHz: E4424B, E4435B 3GHz and 4GHz (E4425B, E4426B, E4436B, E4437B) drop-in version will come a bit later.? Current version can be used with these models but requires a few modification on the Agilent synthesizer PCB. Here are a few pictures of the current version:  If you are interested in getting one of these please contact me off the list.? The ballpark figure for assembled PCB is ?100.? If this looks like a daunting process, for a small fee I can install it into a synthesizer module and test it afterwards if you ship it over. This is experimental work so please consider it as it is.? From my phase noise and performance measurements this looks identical or better than 1GC1-4210. Cheers Leo
|
Amazing work there!
toggle quoted message
Show quoted text
On 11/23/2018 3:27 PM, Leo Bodnar wrote: Update: I have made a successful prototype run of a 1GC1-4210 drop-in replacement PCB for the following instruments:
250 kHz – 1 GHz: E4423B, E4434B
250 kHz – 2 GHz: E4424B, E4435B
3GHz and 4GHz (E4425B, E4426B, E4436B, E4437B) drop-in version will come a bit later.
Current version can be used with these models but requires a few modification on the Agilent synthesizer PCB.
Here are a few pictures of the current version:
If you are interested in getting one of these please contact me off the list.? The ballpark figure for assembled PCB is ?100.
If this looks like a daunting process, for a small fee I can install it into a synthesizer module and test it afterwards if you ship it over.
This is experimental work so please consider it as it is.? From my phase noise and performance measurements this looks identical or better than 1GC1-4210.
Cheers
Leo
|
Very impressive work. Lucky you to have an FSIQ also.
Regards,
Mark
On Fri, Nov 23, 2018 at 1:27 PM Leo Bodnar < leo@...> wrote: Update: I have made a successful prototype run of a 1GC1-4210 drop-in replacement PCB for the following instruments:
250 kHz – 1 GHz: E4423B, E4434B
250 kHz – 2 GHz: E4424B, E4435B
3GHz and 4GHz (E4425B, E4426B, E4436B, E4437B) drop-in version will come a bit later.?
Current version can be used with these models but requires a few modification on the Agilent synthesizer PCB.
Here are a few pictures of the current version:
If you are interested in getting one of these please contact me off the list.? The ballpark figure for assembled PCB is ?100.?
If this looks like a daunting process, for a small fee I can install it into a synthesizer module and test it afterwards if you ship it over.
This is experimental work so please consider it as it is.? From my phase noise and performance measurements this looks identical or better than 1GC1-4210.
Cheers
Leo
|
Hello,
to my maximum embarrassment and as someone who smells an
Instagram photo opportunity: what are you selling here?
Tam
With best regards
Tam Hanna
---
Enjoy electronics? Join 13700 other followers by visiting the Crazy Electronics Lab at
On 23. 11. 2018 21:27, Leo Bodnar
wrote:
toggle quoted message
Show quoted text
Update: I have made a successful prototype run of a 1GC1-4210
drop-in replacement PCB for the following instruments:
250 kHz – 1 GHz: E4423B, E4434B
250 kHz – 2 GHz: E4424B, E4435B
3GHz and 4GHz (E4425B, E4426B, E4436B, E4437B) drop-in version
will come a bit later.?
Current version can be used with these models but requires a few
modification on the Agilent synthesizer PCB.
Here are a few pictures of the current version:
If you are interested in getting one of these please contact me
off the list.? The ballpark figure for assembled PCB is ?100.?
If this looks like a daunting process, for a small fee I can
install it into a synthesizer module and test it afterwards if you
ship it over.
This is experimental work so please consider it as it is.? From my
phase noise and performance measurements this looks identical or
better than 1GC1-4210.
Cheers
Leo
|
The E4435B contains a divide by N hybrid that is very likely
to fail at any instant.
Leo designed a drop-in replacement PCB, to scratch his own itch,
and is making his work available to others with the same problem.
A friendly little gesture that happens quite a lot on these groups.
-Chuck Harris
Tam Hanna wrote:
toggle quoted message
Show quoted text
Hello,
to my maximum embarrassment and as someone who smells an Instagram photo
opportunity: what are you selling here?
Tam
With best regards
Tam Hanna
---
Enjoy electronics? Join 13700 other followers by visiting the Crazy Electronics Lab at
On 23. 11. 2018 21:27, Leo Bodnar wrote:
Update: I have made a successful prototype run of a 1GC1-4210 drop-in
replacement PCB for the following instruments:
250 kHz – 1 GHz: E4423B, E4434B
250 kHz – 2 GHz: E4424B, E4435B
3GHz and 4GHz (E4425B, E4426B, E4436B, E4437B) drop-in version will
come a bit later.
Current version can be used with these models but requires a few
modification on the Agilent synthesizer PCB.
Here are a few pictures of the current version:
If you are interested in getting one of these please contact me off
the list. The ballpark figure for assembled PCB is ?100.
If this looks like a daunting process, for a small fee I can install
it into a synthesizer module and test it afterwards if you ship it over.
This is experimental work so please consider it as it is. From my
phase noise and performance measurements this looks identical or
better than 1GC1-4210.
Cheers
Leo
|
Hello,
BIG thanks.
Now all I need is a good picture of the IC, and I will run it for free!!!
Tam
With best regards
Tam Hanna
---
Enjoy electronics? Join 13700 other followers by visiting the Crazy Electronics Lab at
toggle quoted message
Show quoted text
On 24. 11. 2018 14:17, Chuck Harris wrote: The E4435B contains a divide by N hybrid that is very likely
to fail at any instant.
Leo designed a drop-in replacement PCB, to scratch his own itch,
and is making his work available to others with the same problem.
A friendly little gesture that happens quite a lot on these groups.
-Chuck Harris
Tam Hanna wrote:
Hello,
to my maximum embarrassment and as someone who smells an Instagram photo
opportunity: what are you selling here?
Tam
With best regards
Tam Hanna
---
Enjoy electronics? Join 13700 other followers by visiting the Crazy Electronics Lab at
On 23. 11. 2018 21:27, Leo Bodnar wrote:
Update: I have made a successful prototype run of a 1GC1-4210 drop-in
replacement PCB for the following instruments:
250 kHz – 1 GHz: E4423B, E4434B
250 kHz – 2 GHz: E4424B, E4435B
3GHz and 4GHz (E4425B, E4426B, E4436B, E4437B) drop-in version will
come a bit later.
Current version can be used with these models but requires a few
modification on the Agilent synthesizer PCB.
Here are a few pictures of the current version:
If you are interested in getting one of these please contact me off
the list. The ballpark figure for assembled PCB is ?100.
If this looks like a daunting process, for a small fee I can install
it into a synthesizer module and test it afterwards if you ship it over.
This is experimental work so please consider it as it is. From my
phase noise and performance measurements this looks identical or
better than 1GC1-4210.
Cheers
Leo
|
