Cheap WIFI amp Mods for 2.4 (and up?)

[VK3XDK]

Hello everyone,
I have just tried modding a "cheap" 2.4Ghz WIFI amplifier (PA) with great success, others may like to have a go BUT be warned-the soldering is fine!

I purchased this amplifier thought "Ebay" and think it cost me around $40 ? including postage, there are lots available in different wattages. (this is an 8W** one) I think they are available for 3.4 and 5.7Ghz too.

Anyway the first thing i done was some pretesting, first thing I noticed is that I could only get around 4-5W out of it and it needed around 20dbm of drive. (maybe their power meters are "calibrated" differently to mine? )
These amps run bi dirrectional ie. they have about 10to 14db of RX gain-then go into TX when they sense more than about 2dbm of input. They could be used as-is for CW/digital/FM/AM where there is a carrier to key them.

I want to use them "all mode" and I wanted separate TX RX ports so that I can interface it directly to a transverter So I needed to do some MODS

I will skip through things pretty quickly, excuse any gray areas (or bad grammer )

[VK3XDK]

inside the box

The amp has a rf switch on both the input and output, I kept the output switch (antenna side) and removed the input switch. I then jumpered the switch pads for TX (see photo)
The RX section is an amplification device followed by a ceramic type band-pass filter (VERY HANDY) I dont think the NF of these amps are top figures (around 4db?) but for general use I'm not going to complain!
I added an SMA pigtail just after a capacitor (after the filter) for an RX port.

[VK2GOM]

Just reading the words on the box was amusing! Apart from making little sense, I did like the word 'amtemma'

73 - Rob VK2GOM / GW0MOH

[VK3XDK]

The last thing to do is to add hard TX switching.

The unit as-is has a diode (i think) rf detector, once the detected signal gets to the appropriate level an Analog devices IC sends logic levels to the input and output rf switches.
I noticed that there is a test pad right next to the detection circuit and that there is around 0.3v on the pad when the unit goes into TX. So all I needed to do was to apply a separate voltage!

The easiest way i could come up with is to just use a forward biased diode to produce around 0.5V (plenty to switch the unit to TX)

So the final switching circuit comprises of a series 2.2K resistor (to the testpad) then a diode down to ground. SIMPLE!

To switch the unit to TX is that is needed is to apply 8 to 12Vdc to the "RCA" connector on the panel (see photo) (very handy because most of my transverters have a 8V TX rail)

Clear as MUD? Enjoy!

[VK3XDK]

I am still wondering if the "8W" includes the gain of the antenna!

[VK2XAX]

Hi all,

My "8watt" WiFi amp for 2.4Ghz arrived today and I was able to do some tests. As Graham found, they are not 8 watts.

They are barely 3 watts CW.

They have two YP242024's in parallel and their P1db@6V is 34dBm with their real output about 3dBm below that. Max V for these devices is 8v but the amp runs them at 5.8V which drops to 5.6v under load.

Testing the amp I found I could get a maximum of 34dBm out and that output was relatively constant across the band.

More pics and info here

https://photos.app.goo.gl/4mVCWhHF8yfzTSbR9

regards

Tim

[VK7HH]

Hi Tim,
Interesting photos, I notice you have a Spectrian amp too. What power did you manage to get output from this?

I have one on the way in the post to be driven by a SG Labs transverter.

[ZL2WHO]

I had a play a few years back with a lower power unit.
I found quickly that the o/p device was not suitable for 100% carrier having been designed for a Tx/Rx duty cycle.

Recently I purchased a 1.8G cell PA rated to 20 odd watts but at 2.4 could only get around 10w which is not to be sneezed at.
All this from someone who hasn't really mastered 23cm.. but well on the way.

[VK2XAX]

Hi Tim,
Interesting photos, I notice you have a Spectrian amp too. What power did you manage to get output from this?

I have one on the way in the post to be driven by a SG Labs transverter.

Hi Hayden,

I have not had to much luck with the Spectrian yet.

They seem to need a higher drive level than advertised and I'm not convinced that the suggested biasing for them is correct either since it difers quite a bit from the data sheet, on the devices.

I've only managed to get a few watts out so far but now that I have at least 2w to play with I'll try again to see what I get.

regards

Tim

[VK2XAX]

I had a play a few years back with a lower power unit.
I found quickly that the o/p device was not suitable for 100% carrier having been designed for a Tx/Rx duty cycle.

Recently I purchased a 1.8G cell PA rated to 20 odd watts but at 2.4 could only get around 10w which is not to be sneezed at.
All this from someone who hasn't really mastered 23cm.. but well on the way.

Hi Mark,

Isn't experimentation fun

Depending on the device you might be able to snowflake the input and output to make it happier at 2.4Ghz

Got pictures ?

regards

Tim

[VK7HH]

I have not had to much luck with the Spectrian yet.

They seem to need a higher drive level than advertised and I'm not convinced that the suggested biasing for them is correct either since it difers quite a bit from the data sheet, on the devices.



Tim

No worries Tim, what drive level have you been using?
From what I saw on the listing they need 1.25W drive for 75W out at 24V. Power gain is 18dB.

So I think the SG Lab transverter will be a good fit for full output.

[VK2XAX]

No worries Tim, what drive level have you been using?
From what I saw on the listing they need 1.25W drive for 75W out at 24V. Power gain is 18dB.

So I think the SG Lab transverter will be a good fit for full output.

Just did some testing at 2403.1Mhz and with ~1w drive

At 26v and 12A it will put out 45.9 dbm which is about 39 watts

At 13.8v and 10A it puts out 41.6 dBm which is about 14watts

The only time I got ~60w out of it was near its design frequency of 2.3Ghz.

I''ve set the bias on the devices at 500mA each as per the data sheet for them.

Changing the bias to 2A per devices as per Pyro's advice does nothing to the power output.

Initial tests show the signal to be clean on the SA at 500mA per device. I'd have do some more testing to see if there is any measurable difference between the two bias settings.


regards

Tim

[VK7HH]

G'day Tim,
Just some further comments. I gather you're current limiting? The eBay listing says "Required supply voltage is 24-26Vdc at about 17A".

Perhaps see how much power you get out. at 2.4GHz according to the sweep, it should be around 60W saturated. At 13.8V he specifies 12A current for 25W output.

See how you go, my board arrives soon.
Hayden

[VK2XAX]

G'day Tim,
Just some further comments. I gather you're current limiting? The eBay listing says "Required supply voltage is 24-26Vdc at about 17A".

Perhaps see how much power you get out. at 2.4GHz according to the sweep, it should be around 60W saturated. At 13.8V he specifies 12A current for 25W output.

See how you go, my board arrives soon.
Hayden

Hi Hayden,
No current limiting going on here !
Power measurements are to compression. i.e. increase drive until no more comes out.
I'll do some more tests and document with pics etc.
regards
Tim

[VK7HH]

OK excellent Tim,
Look forward to your findings, likewise I'll post mine.

Might be worth starting a new thread for the Spectrian? There are a few out there so might be beneficial.

Hayden

[VK2XAX]

OK excellent Tim,
Look forward to your findings, likewise I'll post mine.

Might be worth starting a new thread for the Spectrian? There are a few out there so might be beneficial.

Hayden

Yes lets!

I know there are others out there and Alf, VK2YAC, 16km from me has just bought one too.

I'll start a thread with a pic of mine and copy my info so far from this thread

regards

Tim

[VK2XAX]

Hi all,
I found another Wifi amp on eBay that is basically the big amp in a smaller box - see pics.

They have two YP242024's in parallel just like the big amp and I've been able to get just on 34dBm out if it.

Just like the big amp you need to add a PTT line. I accomplished that with a tiny feedthru cap and the same mods as the other amp

I bought mine for $34 inc postage, so that makes for a cheap ~2W amp for 2.4Ghz.

I'll be using this little amp as the driver amp for my Spectrian

eBay link here...

https://www.ebay.com.au/itm/Professiona ... 3076730884

cheers

Tim

[VK7HH]

Good work Tim. An easy and cost effective way to get 2W on 2.4.

I've settled on using 12V for my system. I've got 25W out of the Spectrian at 2400 so gotta be happy with that.
Doing mods on my 3.4 panel now.

73
Hayden