"hardware not found" with new USB to RS232 lead.

I have bought 2 new USB to RS232 leads and wish to use them with PICAXE. I have wired them as normal (2 to Txd, 3 to Rxd 5 to GND) but they both will not find the hardware that works fine with my old cable setup.
The loopback test works and debug transmissions are coming back to the screen. Signals look clean and robust on the CRO.
I have also connected the DE9 connector as a Null Modem (RTS to CTS and DSR to DTR & DCD) just in case.
How does the editor know that it has found the hardware?
Please excuse me if this has been covered before but I have not conjured up the right words in the search to find a solution.
Thank-you in anticipation..

Peter Clark
 

inglewoodpete

Senior Member
Welcome to the PICAXE Forum. Your problem is a fairly common one.

I can see that you're not a beginner with digital electronics: you called the serial connector a DE9 (and you have an oscilloscope of some sort)!

The loopback test works and debug transmissions are coming back to the screen.
When you say that debug transmissions work I presume that you have a programmed PICAXE already, sending data to the PE Terminal. This tells me that the data return path is working and the 'Serial In' leg of the circuit is the culprit.

I have bought 2 new USB to RS232 leads and wish to use them with PICAXE. I have wired them as normal (2 to Txd, 3 to Rxd 5 to GND) but they both will not find the hardware that works fine with my old cable setup.
Are the leads different from your previous one/s? In (Windows) device manager, go to the device's Properties>Details Tab>Hardware IDs and compare them with your cable that works. Your new cable may not be able to transmit a <Break> signal, a requirement for interrupting the the PICAXE chip and placing it in programming mode.
Signals look clean and robust on the CRO.
Does the CRO show a long <Break> signal (pulse high) when you attempt to program the chip?
I have also connected the DE9 connector as a Null Modem (RTS to CTS and DSR to DTR & DCD) just in case.
How does the editor know that it has found the hardware?
The hardware signalling should not affect the programming editor's attempts to program the chip, since you say that the loopback test works. As mentioned previously, the programming editor's <Break> 'high' signal puts the PICAXE in programming made and the PICAXE sends pulses (data?) back to the PE on the return leg. You should see this on your oscilloscope if it has two channels.
 
Thanks for your help Pete,
I'm 72 and been designing and building electronics gear all my working life. Always had someone else programming for me. My main business was high speed PCB design. Fixed a junked HP 54510A 100MHz storage CRO that I would have sold my first born for 40 years ago! The Picaxe was a good way to ease my way into making readily configurable hardware using basic basic.
The good cable starts the programming process with a long1.6 sec. break but the new one shows no activity at all when I should see a break.
I bought two USB- 232 cables on Ebay and neither work.
Is this "no break" capability fixable or do we just have to take our chances with these things as most places that sell them would have absolutely no idea what we are talking about.
 

hippy

Technical Support
Staff member
Is this "no break" capability fixable
Not that I know of. It comes down to the silicon in the USB-to-serial chip used in the cable; if it doesn't support break signalling there is no way to have that and the PICAXE programming cable needs that to work.

There may be ways to implement manually triggered break signalling but it's a lot more complicated than using a cable which does support it.
 
Thanks Pete,
At least I know now what the programming process is.
I will try my luck with other sources. They are quite cheap on Wish or Alibaba.
Thanks again for your help.

Peter
 

lbenson

Senior Member
Not an immediate solution to your problem, but what I use is here. Around $6.50USD for 5 at JLCPCB with slow boat shipping. Suggested by Goeytex, similar used by Erco. Makes programming easy with a cheap CH340 USB/serial module.
 

AllyCat

Senior Member
Hi,

The majority of "named" RS232 USB adapter chips should support break signalling, but there is the further issue of "fake" products. FTDI devices are of particular interest because they can be (software) programmed to invert their signals (to "RS232" rather than the normal "RS232-TTL" polarity) as required for PICaxe programming, but a "cheap" FTDI device almost certainly will be a fake :( . Goeytex is the expert on this topic, for example the following two posts (and surrounding threads) may be informative: HERE and HERE.

Personally, I still use some ancient "Blue" CH340 USB-RS232 (9-pin) adapters, which generally have RS232 polarity but (only) TTL levels (i.e. 0 - 5v., ideal for PICaxe programming). These have been long-promoted as a simple, low-cost option by forum member "manuka", but I cannot vouch for any recently-sold (i.e. potentially fake or "updated") devices.

Cheers, Alan.
 

hippy

Technical Support
Staff member
Maybe try your luck with an inverter first?
Peter describes the cables as USB to RS232, and seemingly with a 9-way D connector, so the output voltage and polarity should be correct for download. It does seem a lack of break signalling is the culprit here.

Most cables do seem to support break signalling but a few do not. Of all the cables I've acquired over the years I have only a couple which do not. It seems Peter was unlucky.
 

lbenson

Senior Member
Personally, I still use some ancient "Blue" CH340 USB-RS232 (9-pin) adapters, which generally have RS232 polarity but (only) TTL levels (i.e. 0 - 5v., ideal for PICaxe programming). These have been long-promoted as a simple, low-cost option by forum member "manuka", but I cannot vouch for any recently-sold (i.e. potentially fake or "updated") devices.
I used those for many years and have several which work, but quit buying them when the last half-dozen or so I bought over the course of several years didn't work.
 

hippy

Technical Support
Staff member
I used those for many years and have several which work, but quit buying them when the last half-dozen or so I bought over the course of several years didn't work.
That's one of the reasons we recommend the AXE027 and don't offer the same support for alternatives. It's often impossible to say what will be in those alternatives or if they will work, especially when they all look the same but aren't. Even when someone recommends a particular alternative, can honestly say they have been using them for years without problem, it doesn't mean that will be the case for others.

I have some cables which are explicitly labelled as "CH340" (or similar) but report themselves to actually be a different chip.
 

lbenson

Senior Member
But the "gold" ch340 usb/serial modules in conjunction with the Goeytex-derived inverter always works for me--I have several dozen of those ch340s in varying usages. Wouldn't buy a cable, though.
 
Thank you all for your information.
If I had just completed the serial cable test instead of stopping at the loopback, I would have found that the static DC test did not work.
We live and learn... eventually.
All the best .. Peter
 
Top