CMU200 Rx level issue / failed loop test

[_Wim_]

I fitted the small 836Mhz filters from Arrow,  part No B39841B4147U410. They work fine.

That good to hear. Hopefully they will keep this stock a little longer.

[450bush]

Dears,

I replaced the 836MHz filter with the B39841B4147U410 and it worked well! My CMU200 now pass the Internal Loopback test!

However, I am still getting some issues with the External loopback 1-4/3-2 that fail some frequencies on 1-4 due to a higher attenuation (just a little...) and also on 3-2 showing less attenuation.

I am working on a small PCB to hold the small filters and be a drop in replacement.

Kind regards,
Pedro

[DavidAzulay5000]

Hey friends
i need help
I need to replace on the RXTX board
the Filter SAW FILTER 836.5MHZ
which was destroyed
At first there was corrosion
I made a temporary solution of coating with copper foil
That temporarily solved the problem
Eventually the filter cracked
I removed the filter using my hot air station
of
hakko fr801
I am now looking for candles to purchase a replacement filter
I found it on eBay
Could this be a decent replacement?

B39841B4122U410 836MHz SAW Filter SMD

https://www.ebay.com/itm/293876605853

Greetings David

[_Wim_]

It will not be easy to fit that without an adaptor board, as that is a TINY component. However, with my limited RF-experience it seems to me that this would at least perform as good as B4147 part from Arrow posted above (and implemented successfully), so worth a try.

Arrow has still almost 18.000 pieces of the B4147 in stock (for only 0.45€ a piece), so that might also be an option.

[DavidAzulay5000]

I know they tiny
But I ran out of options.
It's the only one I found and what's in ARROW
Exactly the same size 3 mm x 3 mm
I have a lot of adapters
I use them for V REG
of linear technology
They are really tiny LT3042
So at most I soldered the filter on the adapter PCB

An example of adapters
This is a multimeter for resistors with values of 5 milli OHM to 10 ohm
that I designed
And built a long time ago

[DavidAzulay5000]

The mission fails
I changed the filter twice
B39841B4122U410 836MHz SAW Filter SMD

In addition, 2 SMD resistors accidentally flew that I de soldered using a hot air station
I looked at the pictures I took before
I'm really mad at myself for not taking a clean picture of this area
From what I've seen resistors are
100 OHM
1 KOHM
The filter I ordered does not pass the internal test
I will try to reverse the polarities
I have no idea what input is
and output
of the circuit
I'm really starting to hate R&S
There is no schematic 

[DavidAzulay5000]

The mission fails
I changed the filter twice
B39841B4122U410 836MHz SAW Filter SMD

In addition, 2 SMD resistors accidentally flew that I de soldered using a hot air station
I looked at the pictures I took before
I'm really mad at myself for not taking a clean picture of this area
From what I've seen resistors are
100 OHM
1 KOHM
The filter I ordered does not pass the internal test
I will try to reverse the polarities
I have no idea what input is
and output
of the circuit
I'm really starting to hate R&S
There is no schematic 

[DavidAzulay5000]

 

[_Wim_]

Hi, where did the resistor come from in the attached picture?

[_Wim_]

And maybe a few other questions: do you measure any resistance between pin 2 and pin 5 of the filter? How does that resistance compare to a brand new filter?

[Ringmodulator]

Hi,

my unit seems to have the same poblems.

Can someone point me to a source for these filters, which delivers to Germany?

Chris

[_Wim_]

Hi,

my unit seems to have the same poblems.

Can someone point me to a source for these filters, which delivers to Germany?

Chris

For the 836MHz part, Arrow electronics as listed above: https://www.arrow.de/products/search?q=B39841B4147U410++&r=true

[DavidAzulay5000]

progress report
The filter 1842.5MHZ
Works perfectly
B39182-B4139-U410 1842.5MHz SAW Filter 50 Ohm SMD

In addition there was a problem with the 9DB attenuation filter
So I made a filter from three size 0402 resistors
t filter
2 24 ohms
1 47ohm
In addition the RF SPDT SWITCH
UPG2179TB
was broken
So I looked for a replacement in other devices
I found the component on the WIFI board
of XBOX 360 E
I removed it and replaced it
The problem remained in the 836.5MHZ filter from the Data Sheet

I think the problem is that the attenuation is too high
at frequencies from 1200 MHz to 2000 MHz
The TYP attenuation
It is over 50DB
And that's why there is a failed internal loop test
Frequency 1200 to 2200
All other frequencies with an accuracy of 0.3 dB
the 1842.5mhz filter of epcos is perfect
B39182-B4139-U410 1842.5MHz SAW Filter 50 Ohm SMD

I do not recommend purchasing the EPCOS 836.5MHZ filter
model B39841-B4122-U410 836MHz SAW Filter SMD
I tried to replace about 4 units with the input and output polarity changed
Exactly the same results in all the tested filters regardless of the polarity of the filter
I thought the minimum attenuation . Maybe you will fall on one of the filters being tested
But they were all measured the same.

[_Wim_]

The problem remained in the 836.5MHZ filter from the Data Sheet

I think the problem is that the attenuation is too high
at frequencies from 1200 MHz to 2000 MHz
The TYP attenuation
It is over 50DB
And that's why there is a failed internal loop test
Frequency 1200 to 2200

Not sure if I misunderstood you, but the goal of this filter is to ONLY allow 836.5MHz to get through (the fixed frequency filter in the attached block diagram), all other frequencies must be attenuated as much as possible. So attenuation can only be too high at 836.5MHz, at other frequencies, it must be as high as possible.

(block diagram copied from here: https://electricalworkbook.com/spectrum-analyzer/)

[_Wim_]

progress report

Do you have an SDR dongle (or other spectrum analyzer) to check if something arrives at the filter (like I have done here: https://www.eevblog.com/forum/repair/cmu200-rx-level-issue-failed-loop-test/msg3917207/#msg3917207)?

My best guess is some RF-amplifier or RF-switch if broken upstream, certainly as you already had to replace another switch in the unit.

[DavidAzulay5000]

Today I checked all frequencies between 1200 and 2200 MHz
are bad
Example with I fed from RF3 frequency generator
In the example -20DB
The signal is low -50DB
Up to frequency 1200 and above frequency 2210
The device works perfectly
I don't have another spectrum analyzer
for this moment i give up 

When I ordered the filters
I ordered 3 types
the 836.5MHZ B122
and 2 types of 1842.5MHZ

the
[50pcs] B69813-N1847-X475 PCN RX Filter 1842MHz SMD
It was really rubbish
installed was no signal at all frequencies
that I installed the
B39182-B4139-U410 1842.5MHz SAW Filter 50 Ohm SMD
All frequencies up to 1200 and above 2200 were perfectly fine again.

With the new filter of 1842.5 The Internal loop test
range of 10 MHz and 1200MHz perfect

range of 2205-2700MHz perfect
succeeded
Better than the original.

The frequency group 1200 to 2200 is really bad
the fm modulation calibration pass
But still failed L1 CO PRO LEVEL CAL

The level is too low. Check the if signal connection

[_Wim_]

I don't have another spectrum analyzer
for this moment i give up 

As an alternative, you could just bridge the filter (IN directly connected to OUT). This will allow the spurs to pass also, but should give you a slightly too strong signal (original filter had a 3dB insertion loss, so I expect the signal to be +3db of what was expected).

If the signal is still to low with a shorted filter, the issue is not the filter but somewhere else. The signal path is relatively easy to follow on these boards (see also my post https://www.eevblog.com/forum/repair/cmu200-rx-level-issue-failed-loop-test/msg3916070/#msg3916070), so than it is a matter to track down where the signal stops (but this does require a way of measuring an RF signal, like a spectrum analyzer or RF-power meter)

[Andrejka]

Hello,

Sorry for offtopic. Maybe someone have additional 1842.5 MHz filters to sell?
 
Thanks
Andrej

[st0ne]

Hi all 

i fixed my CMU200 by cleaning the filters and covering them with conductive paint which is used for repairing keyboard traces!
Maybe this information is helpful 

https://www.amazon.de/gp/product/B07WJTYXPH/

this conductive paint is very fluid and gets hard with a hot air gun and about 100C temperature.

all tests passed

Philipp

[_Wim_]

Hi all 

i fixed my CMU200 by cleaning the filters and covering them with conductive paint which is used for repairing keyboard traces!
Maybe this information is helpful 

https://www.amazon.de/gp/product/B07WJTYXPH/

this conductive paint is very fluid and gets hard with a hot air gun and about 100C temperature.

all tests passed

Philipp

That seems like a very good way to fix this. Much easier than locating the required parts. Thanks for posting!

[Bicurico]

Can you please elaborate?

Am I understnading it correctly, that the cause of error is the oxidation of the shield of the filter and by applying this conductive paint, you are restoring the lack of shielding?

Stupid question: what about cutting aluminium foil to the right size and glueing it on top of the filters? Wouldn't that work, too?

[_Wim_]

Can you please elaborate?

Am I understnading it correctly, that the cause of error is the oxidation of the shield of the filter and by applying this conductive paint, you are restoring the lack of shielding?

Stupid question: what about cutting aluminium foil to the right size and glueing it on top of the filters? Wouldn't that work, too?

See the posts in the beginning of the thread, I have tried this with copper foil, and while it certainly improved the behavior, it was not fully fixed like, so the method Philipp used seems to be better. 

[Bicurico]

Thank you and sorry - somehow I missed this whole thread before, confusing it with other CMU200 repair threads.

This is interesting, because my devices have the same symptoms.