What I meant was... 

Steve, your test is perfect. This is a very good follow-up to the previous video. IMO, the reason why the power output is slow to rise is due to heat within the MOSFET finals. MOSFETs run cool when they are 100% on or used as switches, but they run hot when used in linear mode. There clearly is a design flaw in the 6200. The correct design will decrease bias voltage on the gate as temperature rises to keep the FET from going into "thermal runaway". It looks like either the bias is set incorrectly OR the bias tracking / temp sense feedback circuit is inoperative or wired wrong. It's very much akin to AGC or a governor in an engine. The best way to test and isolate this issue is to measure the drive RF going into the gates and confirm that it is stable. Next I would look with an IR camera and see if there is good thermal conduction from MOSFET's to heat sink and if they are balanced thermally. It could become a deep dive very easily and we would need a full bench with test gear to do the analysis. Keep up the good work. 73 OM

That is very odd behavior

Steve, you are very awesome! Thanks for this test.

Now it all makes sense to me, nice work!

Glad I didn't even consider the 6200 after selling my 6100. Happy with the G90 so far.

Interesting stuff and Cool setup by the way

just a FYI, in the last video clip your audio start going only to the left channel. I usually listen with one earbud in so had to switch to hear it. Thanks for the open and honest reviews.

In before someone figures out how to modify the finals to fix the problem, and then Xeigu makes a hardware revision. I might consider getting one of these after the dust settles. I never got onboard the X6100 train and this seems like a fun radio to play around with. Maybe it's not a fantastic radio currently, but as a hackable Linux machine that is also a radio it's very cool.

At least they didn't lie about the 8W power. I hope they've also fixed the poor WiFi performance and Bluetooth CAT control. I kind of regret buying the 6100. Thanks for the honest review, sir!

Whats sad is you made it clear in the first video. The video was clear even more so because you also included the watt meter in the video.

I thought you were very clear on the last video. Thanks for testing!

My guess... Xiegu's SWR foldback circuit has a high derivative response, and might need a cap value adjusted (down, probably). Would LOVE to see the schematics. 🙂

Thanks to put al the work in this for the Hams that was waiting a long time for this radio ,including me. Now i wait what is the next step from Xiegu. Otherwise i go for the 6100.

Great information on your videos! What dummy load are you using?

This is a bit disturbing with a radio that is supposed to be an "upgrade" to the X6100. I was fully prepared to write this off as a prototype unit having a funky output but then you pulled the card of this is the production model. DoH! That really solidifies that I am NOT getting one of these (although that mic is NICE and I still wonder if it works on the 6100).

Thanks for the explanation and demo! What does this mean in terms of usability? Does the initial ramp up time affect the ability to make contacts?

I remember reading on the Sbitx forum about a problem when keying the radio there were many people blowing internal components because of too fast of reaction switching between receive and transmit (both active at same time for fraction of a second). Then some changes were made to circuitry (and possibly to firmware also) that would slow all that down so there was more time for the circuit to make the switch (because it does not make use of a relay). It also uses RaspPi to control the radio. I wonder if this is a similar situation and perhaps that power up response can be adjusted in software.

Just adjust CW speed so the dits are 5 seconds long.

You proof your point. Strange the X6200 needs that amount of time to deliver full power. The 705 shows your meter reacts like it should. 73 Phil

Well you can't say the Xiegu has an ALC spike. It'd be curious to see what the predriver/PA bias voltage is doing. I understood what you were trying to say in the first video. Good catch. 👍

On my production run X6200, I find that when I'm keying at 20 WPM--using the same SWR meter as you--I'm reading about 6.2-6.8 watts on average (it floats between those) when the X6200 is set to 8 watts output. Thomas (K4SWL)

Scope the output and see if there are any spikes as well.

All this must be very confusing to new hams!

That will be a problem if using some amplifiers. Hopefully that is able to be resolved in firmware (thermal tracking and compensation).

My guess would be either a component isn't the correct value or type, or the firmware has a bad bit of code in it. Something is slowing the power delivery, it's there, but dragging it's heels...

No bueno! I'll wait....or fork out the cash for the 705 if I ever go this direction. Still getting used to an FT-DX10 :)

That’s really odd, I know it’s a RF sensing meter but it kind of resembles DC charging a capacitor, not sure that will be fixable with a new firmware update?

The 6100 doesn't have that issue, either, correct? And.....you're showing that now while I type...I love how you answer my questions in pre-recorded fashion...lol

The next question is does it do the same in other modes. Also, what does the signal sound like when doing CW? Is there a chirp or bad tone?

Yeah, peak versus average metering means nothing for long key down manual CW. It is a weird effect. I just watched K4SWL's video, and it does seem like CW performance is an after thought for Xiegu.

FYI, at 4min 57sec the video's audio when from stereo to left mono only.

that interesting . only the 6200 does this. 73

I’m wondering if this might be a product of the transistors used for the finals… Just my guess … wondering

Here's a REALLY bad hypothesis....it's an SDR, which often limit output based on SWR. Since that is a function of ohmic resistance, could it possibly be lazy code evaluating antenna vs internal resistance and ramping power during evaluation (interenal vs antenna)? I'm not smart enough to answer that question....

Protective circuit... Pray if a firmware update like normal for the brand... I can see it now "Whats the best Firmware" for Power and still have the filtering and Best USB and DATA modes... hehehehe N3WER

The G90 is great, why can’t they figure out these?!

The 1st video was clear, even for someone like me with mush for brains.

Sum thing is telling me they didn’t bias the transistor’s properly ?? May want to check that ??

This is what put me off in your first video Steve but this just clarifies the issue on another level. I could be wrong but i doubt this could be fixed with any firmware update. I guess calling DX 5 times would work haha. Thanks

Definitely not the meter averaging the power. Great test. That’s a weird issue. 73 de K2CJB

Great catch, If the 6100 works, it's my guess, Xiegu can easily correct the problem😏

I would be afraid that I wouldn't get to full power before I had completed the transmission. If they are using cheaper parts that the 705 they may want to slow down the rise to full power so that the output is not over driven. just a thought. 73

So, That means the radio won't be to good on digital modes ( 10 to 20 sec Transmit) What you think?

I wonder if this is something that can be fixed in software. It's definitely an issue... Especially with CW.

Wondering what it is going to sound like on air .... QSB built in OM HiHi. BTW you have found a bug. Thinking it's in there fold back circuit for heat protection. Try it in a very cold room i.e. stick in a cooler full of cold beer.

I wonder if it's an ALC issue. I wonder if it does the same thing on FM mode

Lets just call it a feature and move on!!

I would be interested to see if Xiegu have any response to this test. This is definitely not desirable behaviour, I think!

Hopefully this can be resolved with a firmware update

I wonder if this will create a "chip" on CW

Very weird. i have the 6100. I see absolutely no reason to upgrade. Less power, slower warmup on transmit, no expandability...

X6200 has a defective feature and not ready for prime time. Should we be surprised...?

If I were Xiegu, I would be as transparent as possible with Beta testers like you. Treating this as a 'black box' doesn't help the cause.

Yep, it’s a bucket of sh….

I dont see anything different here, you CAN'T compare Icom quality with Crapy Xiegu quality duhhh. For a reason is way cheaper