jcheek
Expert Member
Update: B593 "Antenna detection" (aka the dreaded "DC short" issue)
So, a quick update/correction to this previous post and the ones that followed from it.
Yesterday I was doing some work with a different multimeter and decided to re-test the flat whip antennas included with my B593s-601. The previous test (done with a Fluke 27 DMM) had indicated a nominal 1.2kΩ DC resistance between the antenna terminals, possibly shunted by two anti-parallel diodes giving 0.6V drop in either direction.
After re-testing with both a Fluke 289 and another meter, I'm now pretty sure that what's inside is just a plain 1.2kΩ resistor. I was alerted to this by the fact that the Fluke 289 showed the same fixed 1.2kΩ resistance in both directions, but also indicated 1.2V (in either direction) on the diode test range. 1.2V and 1.2kΩ ? Hmm, sounds like 1mA, and it was.
Turns out that on its "Diode Test" range, the Fluke 289 uses a 1mA test current (1.0mA x 1.2kΩ = 1.2V) whereas the Fluke 27 uses 0.5mA (0.5mA x 1.2kΩ = 0.6V). The 0.6V was just coincidence that happened to look like one diode drop.
So, without actually opening up the antennas up or using a fancy network analyzer, I'm now pretty convinced that what's inside the B593s-601's flat white whip antennas (for providing DC continuity) is just a plain old 1.2kΩ resistor.
Providing DC continuity in an antenna (either through a short, or a deliberate resistance value), and then using a DC bias voltage on the antenna port to test whether the antenna is actually present, is a pretty straightforward technique. In automotive applications, it can also be used to detect abnormal antenna conditions (eg ground faults and short-circuits). From what I've read on the topic, 10kΩ is quite a common value to use inside such an antenna, but there's nothing stopping Huawei using a different value.
(For those interested in more detail on how the technique works, read this and this).
A quick check on the B593's SMA antenna ports also reveals that both have a constant 1.8VDC potential across them (interestingly, the centre pin is negative with respect to the outer shell) :
This makes the case for antenna detection by DC resistance measurement on the B593 family that much stronger.
So, what does all this mean for the average B593 user who wants to use an external antenna and can't get the device to detect it ? Well, if you own a B593s-601, you probably don't care since you can force the modem to use the external antenna even if the "Auto" setting doesn't work.
For people who have the B593u-devices (with only auto antenna detection), and antennas that don't have DC continuity, this might mean that you can force the B593 to detect your antenna by providing DC continuity across the antenna terminals at 1.2kΩ (or perhaps less). Technically, this is much less of a challenge than trying to provide a DC short on an antenna that doesn't have one (without messing up it's RF performance, of course). This might well be what is being referred to in other Web articles on the topic, like this one :
"Tele2, on the other hand, never reacted to the problem and has not offered any new firmware. This operator uses LTE 900 and 2600. The only remedy are the adapters which contain a small resistor (Widerstand) and thereby makes the router shift automatically to external antennas"
(Thanks jeaninek for the original link to that thread).
I would still caution that forcing the B593 to use an external antenna obviously won't improve the signal conditions if the antenna itself is unsuitable or not set up right.
Also bear in mind that all of my tests have been done on a B593s-601. Whilst it's possible (likely, even), I can't say for sure whether the same things are true for the B593u-12 or B593u-91 and its accessories.
So, a quick update/correction to this previous post and the ones that followed from it.
Yesterday I was doing some work with a different multimeter and decided to re-test the flat whip antennas included with my B593s-601. The previous test (done with a Fluke 27 DMM) had indicated a nominal 1.2kΩ DC resistance between the antenna terminals, possibly shunted by two anti-parallel diodes giving 0.6V drop in either direction.
After re-testing with both a Fluke 289 and another meter, I'm now pretty sure that what's inside is just a plain 1.2kΩ resistor. I was alerted to this by the fact that the Fluke 289 showed the same fixed 1.2kΩ resistance in both directions, but also indicated 1.2V (in either direction) on the diode test range. 1.2V and 1.2kΩ ? Hmm, sounds like 1mA, and it was.
Turns out that on its "Diode Test" range, the Fluke 289 uses a 1mA test current (1.0mA x 1.2kΩ = 1.2V) whereas the Fluke 27 uses 0.5mA (0.5mA x 1.2kΩ = 0.6V). The 0.6V was just coincidence that happened to look like one diode drop.
So, without actually opening up the antennas up or using a fancy network analyzer, I'm now pretty convinced that what's inside the B593s-601's flat white whip antennas (for providing DC continuity) is just a plain old 1.2kΩ resistor.
Providing DC continuity in an antenna (either through a short, or a deliberate resistance value), and then using a DC bias voltage on the antenna port to test whether the antenna is actually present, is a pretty straightforward technique. In automotive applications, it can also be used to detect abnormal antenna conditions (eg ground faults and short-circuits). From what I've read on the topic, 10kΩ is quite a common value to use inside such an antenna, but there's nothing stopping Huawei using a different value.
(For those interested in more detail on how the technique works, read this and this).
A quick check on the B593's SMA antenna ports also reveals that both have a constant 1.8VDC potential across them (interestingly, the centre pin is negative with respect to the outer shell) :
This makes the case for antenna detection by DC resistance measurement on the B593 family that much stronger.
So, what does all this mean for the average B593 user who wants to use an external antenna and can't get the device to detect it ? Well, if you own a B593s-601, you probably don't care since you can force the modem to use the external antenna even if the "Auto" setting doesn't work.
For people who have the B593u-devices (with only auto antenna detection), and antennas that don't have DC continuity, this might mean that you can force the B593 to detect your antenna by providing DC continuity across the antenna terminals at 1.2kΩ (or perhaps less). Technically, this is much less of a challenge than trying to provide a DC short on an antenna that doesn't have one (without messing up it's RF performance, of course). This might well be what is being referred to in other Web articles on the topic, like this one :
"Tele2, on the other hand, never reacted to the problem and has not offered any new firmware. This operator uses LTE 900 and 2600. The only remedy are the adapters which contain a small resistor (Widerstand) and thereby makes the router shift automatically to external antennas"
(Thanks jeaninek for the original link to that thread).
I would still caution that forcing the B593 to use an external antenna obviously won't improve the signal conditions if the antenna itself is unsuitable or not set up right.
Also bear in mind that all of my tests have been done on a B593s-601. Whilst it's possible (likely, even), I can't say for sure whether the same things are true for the B593u-12 or B593u-91 and its accessories.
Last edited:
