How can I improve cw keying with Flex-1500?

Phil N3ZP

Running Windows 7 on HP DV6 laptop, 4GB memory, 4 USB ports, wireless card off. CW keying is very choppy even using external keyer. Cannot seem to find optimum delay setting.

Tim - W4TME

First let me explain the technical issues that can cause the delay (latency) you are experiencing. All of the radio's "command and control" data is being sent down the USB interface. USB is a polled interface, meaning that the PC's clock is used to attempt to request and send data at a steady rate. Turning the transmitter on and off is one component of the command and control data. If there are processes going on inside the PC that causes system latency, then the "steady rate" isn't so steady resulting in variable latency. Since CW is a very strict timing mode, this latency shows up as a delay in sidetone if using the internal keyer or choppy sending if using an external keyer. Unfortunately, the speed of your processor has little do do with this variable latency issue. It is a problem that is dependent on the hardware interfaces integrated into your PC and their associated drivers. (Hint: update all of your hardware drivers and PC BIOS) Another issue is laptops vs. desktops. Laptops are not engineered for throughput performance and low internal system latency. They are compromised engineering designs to reduce weight and get long battery life. Desktops are much better suited for low latency and high throughput performance. So what can you do? First off, in addition to turning off the wireless, turn off the Bluetooth too. Make sure that your FLEX-1500 is the only device on the USB controller it is plugged into. See the following article. (http://helpdesk.flexradio.com/entries/22821743-FLEX-1500-USB-Performance-Degradation-Due-to-Other-USB-Connected-Devices) You also want to check to see what your DPC latency looks like. (here is an explanation of what DPCs are: http://helpdesk.flexradio.com/entries/23167502-What-are-DPCs-and-why-do-they-matter-). Here is an article on how to test to see what your actual DPC latency numbers are for your PC while running the radio (http://helpdesk.flexradio.com/entries/21901482-Using-LatencyMon-to-collect-DPC-Latency-Data) If the DPCs are very high (any event >2500) then you are going to have challenges with that particular PC running a real-time audio processing application like PowerSDR.

Phil N3ZP

Hi Tim, Thanks for the information. I do use a wireless USB mouse rather than the pad on the laptop but did run the software to see which buss it's on have it on a separate buss. I did update BIOS a while ago but have not cheked drivers recently. I will check the latency. I do have a couple of dual core Intel MacBooks with Windows 7 running in Bootcamp, but have not tried them recently. I also have a Windows XP desktop with a 2.8 GHz P4 and 3Gb of memory. I use it for most of my ham activity and have it connected to a K3 via serial port. It has a separate 4 port USB card in addition to the built-in ports. It's a slower machine but can I optimize it in any way? I do intend to upgrade the desptop but not for a few months. Thanks, Phil

Rob N4GA

I completely solved this delay issue on my FLEX 1500 by simply turning off the VAC when using CW.

In which case, you then get your audio from the phones jack on the front of the 1500 or using FLEX wire audio out on the back.

I operate at 25wpm and there are NO issues. I do not use QSK full break-in. Set the T/R delay to like 600ms.

I never got people that want to work CW full QSK. Maybe receive between each word, but between every dot or dash? Hmmm..

Rob, N4GA

 

Carl Moreschi

Just a comment on why full QSK is a good thing.  When you are working a split DX pileup, full QSK allows you to stop sending the instant the DX station transmits.  If everybody had this, the QRM would go way down in these pileups.

Rob N4GA

Understand, thanks.

However, still not worth it to me. Switching transceivers and amplifiers in and out on every dot or dash just seems like overhead on all systems that is not worth it.

It's a individual thing, I know.

73