|
Thread: E220 & E170 : disconnection problem solved (h/w)
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
E220 & E170 : disconnection problem solved (h/w)
|
|
Posted:
2008-09-14 03:50
|
Reward:
0
|
Thread Level:
General User
|
|
This thread refers to the 'randomly' disconnects of USB broadband modems as E170, E220. The disconnections are like a physical removal of the modem from the USB port. This problem is caused by a h/w reason on the USB power supply, so the solution is an external hardware patch.
Note that I am a user of an E170 modem which is my sole internet connection for business and personal use. I signed a contract with a local mobile broadband provider, they gave me the E170 and ... the problems started!
You can read the full story (which is similar to those others) at my posts:
http://forum.huawei.com/jive4/thread.jspa?threadID=323238
http://forum.huawei.com/jive4/thread.jspa?threadID=321295
Finally I have my problem SOLVED with an external electronic 'patch'. After this fix, I am fully satisfied with my modem and believe that this is a good product. I can be 'always connected browsing or downloading for hours with speeds up to 4Mbits/s (download)' using a small 'patched' cable or a 5 meters 'patched' USB extension cable (not repeater or USB hub). I did not change any driver, application s/w or the modem's firmware.
Recently I contact another user of the forum.huawei.com with a similar h/w disconnecting problem on an E220 modem and we SOLVE it:
http://forum.huawei.com/jive4/thread.jspa?threadID=323536
This h/w patch is to add some capacitors on the USB power line, 'filtering and reserving' DC voltage for the 3G transmitter preventing the digital disconnection from the USB port. This patch must be done by a technician because we don't want to 'short circuit' any cable!
Of course more typical and preferable for the customer is to have this fix at 'no cost' from the PC or the MODEM manufacturer because it is an extreme incompatibility issue at the USB interface. Two months after solving the problem I found an article for electronics engineers (Huawei designers may have read it) with the conclusion ... don't forget the capacitor! Read it at:
http://www.epn-online.com/page/new58312/stay-active-selecting-the-right-passive.html
----------
CONCLUSION (from my point of view and for my own case)
Problem is: reset or 'brown out state' of the modem
Caused by: USB power ripple/spikes when modem transmits a lot of data, or at high power (other possible reason: a USB connector malfunction like tremble, crack, dirt etc.)
H/W Solution by the user: contact an electronics technician to add capacitors externally (on a patched cable) 'filtering' the ripple spikes (first check and clean the USB connectors)
H/W Solution by the PC manufacturer: Do better design on the USB port to supply a 'good & steady' power for new advanced USB peripherals (as USB mobile broadband modems)
H/W Solution by Huawei: Add decoupling capacitors or split the power lines of the CPU, USB I/O and GSM/3G circuits to prevent resets and brown outs at difficult (for the circuit) situations.
S/W Solution at firmware: Sorry NOT available, ... or possibly a new firmware could 'amplitude modulate' the RF-TX signal to smooth the rise and fall of the RF PWR OUT at every data burst (true artificial intelligence), or change the threshold of the reset and brown out detect circuits to lower levels which may degrade the total performance of the circuit.
S/W Solution at driver: Sorry NOT good, ... or a new s/w driver could see the disconnects and try to reconnect and if not succeed would restart the computer, etc. I am not joking. Programmers are trying to use E220 for a 'reliable' application:
http://ubuntuforums.org/showthread.php?t=914024
----------
SERIOUSLY the above could solve at least the following threads (which are carefully determined as 'the same problem'):
SOLVED AT:
- E220 disconnecting
. http://forum.huawei.com/jive4/thread.jspa?threadID=323536
Other case SOLVED by a 'replacement' modem!
- E172 disconnects - massive stability problems
. http://forum.huawei.com/jive4/thread.jspa?threadID=321438
MAY SOLVE:
- E220 looses power?
. http://forum.huawei.com/jive4/thread.jspa?threadID=319534
- E220: the connection aborts unregularily
. http://forum.huawei.com/jive4/thread.jspa?threadID=322459
- E220 Mac woes...
. http://forum.huawei.com/jive4/thread.jspa?threadID=322248
- E170 constantly crashing
. http://forum.huawei.com/jive4/thread.jspa?threadID=321295
- E169 Strange connection drops in Vista
. http://forum.huawei.com/jive4/thread.jspa?threadID=323238
- mobile broadband usb device disconnects
. http://forum.huawei.com/jive4/thread.jspa?threadID=323555
POSSIBLY SOLVES:
- E 172 crash problem on mac os x 10.5.1
. http://forum.huawei.com/jive4/thread.jspa?threadID=322460
- E220 disconnects all the time (post of user 'Prosessor')
. http://forum.huawei.com/jive4/thread.jspa?threadID=321470
Regards
George
P.S.: I cannot offer the cable or the patch, just sharing my experience.
|
|
|
oddyseus
|
|
General User
|
|
Posts:
22
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 17, 2007
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi GeorgeVita,
Thank you for sharing your brilliant idea with us. Is it possible if we replace the H/W inside the PCB rather than using a patched cable? If it is, what component that we should replace? Any help much appreciated...
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi oddyseus,
I would, if I could! I cannot because this 'voids the warranty' of the modem or the laptop and I cannot ask for the schematic of the circuit to "try it". Theese modems (and the notebooks) are advanced and highly integrated devices so any internal modification can be done only by a trained & informed service center.
My idea was to determine the problem and then I thought the external 'patch' on the cable. When providing a 'smoother' power (5V) to the modem it extends the "life" of the device (higher MTBF) and possibly minimizes unwanted parasitic emissions and radiation. Also It runs already 'cooler' than with the disconnects! At the other hand the 3G provider processes a normal "connect-data-disconnect" flow than the "connect-data-connect-connect-connect..." (when you are h/w disconnecting the modem does not send the appropriate disconnect request to the provider).
Regards
George
|
|
|
Christoph
|
|
General User
|
|
Posts:
4
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 26, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi George,
I'd like to try your solution, can you share some details on what type of and how many capacitors you used or even a circuit layout of your patched cable?
Regards,
Christoph
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi Christoph,
at first I would like to remind the typical answer from Huawei stuff x00114004 who advises: "Hi ,please contact the local repair center to check if the E170 has the problem truely ,if it does ,they will help you via another way ,thanks!" at thread http://forum.huawei.com/jive4/thread.jspa?threadID=323748 and to mention again that there is a "solved" case from a repair center or a local provider at thread
http://forum.huawei.com/jive4/thread.jspa?threadID=321438
--------------------------
If you want to try the cable patch you ought to know a little bit of electronics or you must contact a technician.
AMATEUR SOLUTION:
From a standard USB extension cable (1m, 1.5m, 5m tested) at a distance 3-5 cm close to the modem's connector end (mini USB for E220, female USB for E170), take apart power lines (#1 Vbus cable=red +5V and #4 GND color=black ground 0V) and solder ALL IN PARALLEL the following: capacitor 470uF/16V, a second 470uF/16V, tantalum 33uF/10V, ceramic 100nF, ceramic 100pF and also a resistor 15K (to discharge the electrolytic capacitors at power off).
DON'T CUT THE DATA LINES (green and yellow cables) OR THE SHIELD CABLE. You can fit all these into a plastic tube with external heat shrink sleeve. Plastic tube may come from a fax or calculator paper roll. Check that the tube passes over the connector (it passes over E220's mini USB).
BE CAREFUL with polarity of capacitors and DO NOT SHORT CIRCUIT the power lines or the shield from each other.
Possibly it works with less capacitors but you have to check it with more than one notebooks to find the ripple threshold that don't force the modem to reset.
EXPERT SOLUTION:
Work a little bit on your lab to find out the minimum capacitance needed in order to gain space. Construct a small PCB and solder all above using SMD PCB connectors for I/O. The final design could be potted to form an "adaptor like" Power Supply Filter.
Regards,
George
*** P.S. Post the good results after tests! ***
|
|
|
Christoph
|
|
General User
|
|
Posts:
4
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 26, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi George,
thanks for the info, I'll try to find someone with experience in the field to build the cable for me.
Regards,
Christoph
|
|
|
Christoph
|
|
General User
|
|
Posts:
4
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 26, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
...regarding the tantalum 33uF/10V: I was only able to find tantalum 33uF/16V so far, am I just searching in bad shops or could that be a typo?
Regards,
Christoph
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
The working voltage is 5V so you can use any capacitor with a specification over 5V. The 16V series is a little bit bigger in size. Be careful with polarities.
G
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Recently a friend did not (!) solve the problem with my patched cable, so I had to revise it. The result is a smaller and better patched cable!
The new approach is to add a schottky diode in series with V+ (Red cable) that blocks any returning 'power reserve' stored in the capacitors. The result is a smaller patch with 3 components:
a. Schottky diode (MBR340, 1N5817, 1N5822) rated at least 1A, choose that with the minimum Voltage drop (Vf) at 300mA (If). Anode goes to PC and Cathode to modem.
b. Electrolytic capacitor 100uF/16V (possibly works with a smaller one)
c. Tantalum capacitor 3.3uF/16V
The above tested (as always) with my laptop and a 5 meters long simple and cheap extension cable (not "repeater" or "amplifier").
In the following link I have placed some photos of 'how to make it' (fully amateur construction of the first version) where I am going to updated with the 'newer'.
http://www.acomelectronics.com/GeorgeVita
(P.S.: I cannot offer the cable or the patch, just sharing my experience)
|
|
|
GsqGsq
|
|
General User
|
|
Posts:
3
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Nov 13, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi,
Regarding this PCB http://www.acomelectronics.com/GeorgeVita/e170fix_foto/cp_3.jpg is it the newer design?
Are you able to provide a simple drawing of the wiring, I'm interested to try this solution, and would like to get a knowledgeable friend to help.
Thank you.
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi, the newer cable patch uses only 3 components:
a. D1 Schottky diode MBR340. Anode goes to PC and Cathode to modem.
b. C1 Electrolytic capacitor 100uF/16V (+ goes to RED V+ pin 1 of modem connector)
c. Tantalum capacitor 3.3uF/16V (+ goes to RED V+ pin1 of modem connector)
Connections are as in the following draft:
Computer . . . .. . . . . . . . . . . . . . . . .Modem
USB conn. . . . . . . . . . . . . . . . . . . . .USB conn.
Pin No.. . . . . . . . . . . . . . . . . . . . . . .Pin No.
1 --------!>!---------+-----------+------------- 1 V+
. . . . . D1. . . . . . .!. . . . . . . !
. . . . . . . . . . . . ------ . . . .------
. . . . . . . . . .C1 ------ . . . .------ C2
. . . . . . . . . . . . . .! . . . . . . .!
4 ---------------------+-----------+------------- 4 GROUND
2 ------------------------------------------------- 2
3 ------------------------------------------------- 3
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
As the above "schematic" is not readable, you can see a better one at:
http://www.acomelectronics.com/GeorgeVita/e170fix_foto.html
|
|
|
GsqGsq
|
|
General User
|
|
Posts:
3
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Nov 13, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi GeorgeVita,
Thanks for drawing out the diagram, as it is now clearer.
I hope this can solve my problem too.
Thanks!
|
|
|
z00134964
|
|
Huawei Staff
|
|
Posts:
656
|
|
Expert Points:
805
|
|
Useable Points:
10905
|
|
Registered:
Jul 10, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
|
Hi，GeorgeVita I am very interested for your excellent draft.Can you send your clear draft to my email <a href="mailto:zhengzhe@huawei.com">zhengzhe@huawei.com</a> I will check this solution if it really works. Great thanks!
|
|
|
GsqGsq
|
|
General User
|
|
Posts:
3
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Nov 13, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
|
Hi GeorgeVita, I have just re-read all your threads about this problem, you seem to mention the problem of USB physical disconnection (Windows sound). Although I do not actually experience the physical USB disconnection issue, does this solution also fix the problem of the connectivity freezing (so call disconnecting) as usually I run ping and whenever I'm having problems I might get high latency or ping timed out totally and I'm unable to connect to any sites. I suppose it doesn't hurt to try since I wouldn't mind a cooler modem.. haa. I'll feedback once I'm able to get this setup. Thanks.
|
|
|
zakeros
|
|
General User
|
|
Posts:
1
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Dec 6, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi.
@GorgeVita.
I made cable patch according to your schematics - it works perfectly! Thanks!
@GsqGsq
I had a similar problem - sometimes connection is freezing - no PING, nothing.
It happens randomly.
Didn't find any solution - yet.
|
|
|
deciom
|
|
General User
|
|
Posts:
4
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 1, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Thanks, Georgevita you solved my problem:
http://forum.huawei.com/jive4/thread.jspa?messageID=397530
Very, very thanks.
|
|
|
fa_wildchild
|
|
General User
|
|
Posts:
1
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Apr 29, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
thanks GeorgeVita
solved my problem. im using huawei e156g
|
|
|
PatrickD
|
|
General User
|
|
Posts:
1
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Apr 30, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
|
I have had the USB port stop working after the E 270 modem "crashed" on my MacBook, I wonder if anyone else has experienced this"
|
|
|
christos_uoa
|
|
General User
|
|
Posts:
2
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jun 6, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hello George. I have the same problem you met with Huawei 172 from Vodafone Greece. Please could you send me an email to christos_uoa@hotmail.com to ask you some things in Greek about the solution. (Greek names of the parts, stores etc).
Se efxaristw
Christos
|
|
|
christos_uoa
|
|
General User
|
|
Posts:
2
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jun 6, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hello George!
Today i applied your solution and finally i have a stable connection without modem's physical disconnections!!! Thank you a lot!!!
I remind you ppl that my problem was physical disconnection of my USB E172 modem. These disconnections occured during high speeds with long durations. I.e. video streaming in youtube, downloading file from ftp,. radio streaming etc. George's solution with the diode and the 2 capacitors solved my problem
Christos
|
|
|
Mark_G
|
|
General User
|
|
Posts:
2
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 13, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
|
|
VasanRS
|
|
General User
|
|
Posts:
1
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 20, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
|
Hi George! I have a Huawei E220 which i bought from my friend but he didn't gave me the original cable! So i bought a new cable in a shop.. The new cable is about 1.5M long... My real problem is modem gets disconnected oftenly.(within 3minutes) Do u mean the length of the cable affect this problem or what? <img alt="" src="/jive4/scripts/fckeditor/editor/images/smiley/msn/sad_smile.gif" />. Please help me George!!!!!! I believe you are a great technician in Electronics and Computer. If you would like to help me George please send me a cable or give me replies on this! My Home Address Name : Vasan Rajeswaran Address : 62A Quarry Road, Dehiwala, Colombo, Srilanka. Postal Code : CO 10350 Im awiting for a positive reply from you!
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi VasanRS,
I understand your problem, as I also faced in the past, but I cannot offer the cable or the patch in any way other than 'described it'.
Q: Do u mean the length of the cable affect this problem or what?
A: 'Small or No cable' is better than 'long cable'. (with the E220 you are always using a cable)
For your case you must first declare the source of disconnects as many times the problem comes from other reasons as the provider (signal strength, network load).
My solution "refers to the 'randomly' disconnects which are like a physical removal of the modem from the USB port. At windows you can hear the sounds of the USB-peripheral disconnection and reconnection. This problem is caused by a h/w reason on the USB power supply, so the solution is an external hardware patch."
Test the modem to another PC (desktop prefered) and compare behaviour. Once you have determine that the problem is the above, you have to construct the patched cable. Take advice from a technician if you cannot do it by yourself! Note also that some users solved the same problem by using an externally powered USB hub (I did not test it). I am personally using patched cables (70cm, 180cm, 3m) without any active components to interfere.
The simpler circuit consists of 3 components, a Schottky diode MBR340 or equivalent and two capacitors. Schottky diodes exist into any switching power supply so a repair technician can find it (he needs it to repair a power supply or he can 'recycle' an old one). You need an electronics technician not an expert or an engineer .
Regards,
George
|
|
|
dantares
|
|
General User
|
|
Posts:
2
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Aug 1, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi Georges,
Thank you very much for this information
I have a Huawei E160 and I encounter the same problems of disconnection.
I made cable patch according to your schematics with a Schottky diode SB340 and not MBR340.
This did not solve my problem.
Is it imperative to use a diode MBR340 to realize the cable?
Thank you for your reply.
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi dantares ,
I checked the datasheet of SB340 and it is OK. If you are sure about the circuit and this did not solve your problem then the problem could be caused by another reason. The only 'trim' you could do is to raise the capacitance to 1000uF and check that capacitors are from the modem side (after the diode).
Regards,
George
|
|
|
dantares
|
|
General User
|
|
Posts:
2
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Aug 1, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi Georges,
For small cable (Schottky diode MBR340) distance must be 3-5 cm close to the modem's connector end?
Thank you for your reply.
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi dantares,
it is not critical the actual mechanical distance as you do not cut the data lines. I have checked it with various cables and connector assemblies (small PCBs).
The main idea is first to determine that the problem seems to be the same ('randomly disconnects like a physical removal of the modem') and then try to 'smooth' the power in transient or high power demands from the modem. Another cause could be the connector so check it (it must be clean and connect firmly).
Regards,
George
|
|
|
e220user001
|
|
General User
|
|
Posts:
3
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 15, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Yeah- the basic idea *REALLY* works. After ten firmware upgrades, cable changes, re-installations etc. I bought cheap USB Hub from Skymaster, see
http://www.preissuchmaschine.de/preisvergleich/popup.asp?pid=344419784&pnr=1&foto=1
It even HAS positions for capacitors on PCB, but they are not installed. The box can easily be opened, there is enough space. I soldered 220 µ and Tantal with 3,3µ. Since then using tis modified hub - the modem works perfectly ! Yeah. People try it- its simple and brilliant!
Thanks VERY much to GeorgeVita!!!
|
|
|
jarviser
|
|
General User
|
|
Posts:
51
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Aug 2, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
George, Looks good. I have added a link in my own website.
Can I ask (I hated electronics at Uni) what is the purpose of the 3.3mfd tantalum bead capacitor? All it seems to do at basic school physics level is add to the 100mfd capacitor making total of 103.3mfd. There must be a more scientific reason that I don't understand.
Also if the schottky is omitted what are the dangers. Is it protecting the PC USB port? In other words, what may go wrong if I just apply a single 100mfd or 1000mfd capacitor across the 4.5volt lines and nothing else at the dongle end of the lead? Forgive my agricultural approach to electronics
|
|
|
e220user001
|
|
General User
|
|
Posts:
3
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 15, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi Jarviser,
each technology of capacitor has different behaviour. Electrolytic types have high capacity and can buffer large charge fluctuations (e.g. after rectifier in power supply). On the other hand, they are "lazy" to respond to ultra short spikes (electronically speaking, they have inductive portion and take charges slower).
So for high frequency blocking you should add second technology type to the electrolytic capacitor.
The reason for switching electrolytic, tantalum, ceramic and probably even polyester types in parallel to a cluster therefore follows the aspect to treat all eventualities in surge protection. It is not the total capacity that matters here, but their different behaviour related to frequencies.
Thats it to my knowledge, but perhaps GergeVita can and will add something....
|
|
|
jarviser
|
|
General User
|
|
Posts:
51
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Aug 2, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
|
Ah yes, frequency dependant impedence. I forgot all that!
|
|
|
GeorgeVita
|
|
General User
|
|
Posts:
42
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Jul 16, 2008
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi jarviser and e220user001 !
jarviser: 'I have added a link in my own website.'
>>> Thanks for spreading any idea!
jarviser: 'if the schottky is omitted what are the dangers. Is it protecting the PC USB port?'
>>> The main idea is to keep the 'filtered' power for the modem and not for the whole PC via the USB port, as described above: 'The new approach is to add a schottky diode in series with V+ (Red cable) that blocks any returning 'power reserve' stored in the capacitors.'. Without this diode you need larger capacitance to achieve same results (first circuit had over 1000 uF).
e220user001 thanks for your 'exact' description!
Regards,
George
|
|
|
e220user001
|
|
General User
|
|
Posts:
3
|
|
Expert Points:
0
|
|
Useable Points:
10000
|
|
Registered:
Sep 15, 2009
|
|
|
|
Re: E220 & E170 : disconnection problem solved (h/w)
|
Hi George, I just did my tiny part. I am SO happy still. This problem accompanied my computer life almost one year - you have to imagine... It was even such, that when it failed by famous "wireless device does not exist.."., it could only be reanimated by flashing firmware into the device (!)
Yesterday I tried a hardcore session: 11 hours online on my E220 - not a single break. Great :-)
Why on earth dont they integrate some capacity into the modem?? No space? Cost saving?
Btw, in my despair I tried one Huawei E160 some months ago, given by my provider for testing purpose - with the same effect on my laptop, my private PC and the desktop PC of one friend of mine. It is *really* stupid, problem passing over generations of devices...
I will finish my contribution to this topic, not without inviting everybody to tinker the capacitance modification you proposed!
e220user001
|
|
|
|
