Fluke 45 Serial Port - UK Vintage Radio Repair and Restoration Discussion Forum

WaveyDipole · 13th Sep 2017, 5:40 pm

Does anyone have information on the pinout of the serial port in the Fluke 45? The service manual does not seem to mention this although it goes into detail as to how the signal goes from the processor to reach the port and vice versa.

I have tried to connect mine to the PC but cannot get it working. I have tried a straight cable as well as a null-modem one and neither works. If I plug my USB-to-serial adapter in directly, then all segments come on after a couple of seconds (like at startup) and stay on until I unplug the cable. I don't have a computer with a hardware serial port on it.

My older PL2303 converter seems to work with everything I throw at it, but the Fluke port has the opposite gender so I have had to buy another converter with a female plug to connect with it. Since I have nothing else to connect it with to test it, I have the GND, TX and RX pins with a logic analyzer and it is transmitting. Still, I have no way to be sure whether the converter is at fault or a custom pinout has been used on the Fluke.

Any information would be of great help.
Thanks.

duncanlowe · 13th Sep 2017, 7:00 pm

The user manual has information on setting up the port and seems to suggest its a standard RS232 interface rather than something with a custom pinout.

MrBungle · 13th Sep 2017, 7:17 pm

There are a lot of dodgy PL2303 converters out there which are missing level shifters and misbehave with certain serial port implementations.

If you can get it open, there should be a MAX211 or similar shifter on one side and an unmarked or prolific branded SSOP package on the other side.

If it's missing the level shifter or the package is a mysterious epoxy glue blob then the adapter is at fault for certain!

The fake ones don't work with my Uni-T UT61E showing the same symptoms. The real one does.

buggies · 13th Sep 2017, 7:24 pm

Hi,

According to the operating manual (I don't possess that DVM), the pinout is marked on the back panel just above the connector. The drawing shows:-
Pin 2 RX
Pin 3 TX
Pin 4 DTR
Pin 5 GND

The manual gives the method to detect if enabled on page 5-2.
The serial defaults are 9600, 8, N, 1.
It can talk to a dumb printer or terminal so even Windows should be able to cope.

G0HZU_JMR · 13th Sep 2017, 8:49 pm

I'm fairly certain it's a null modem (swapover) cable. The manual also hints that it is as null modem cable as it says to use a cable designed to link two IBM PCs together.

I've got a Fluke 45 here and I have used it a few times over RS-232 and I use it with a USB to RS-232 converter and a commercial (not homemade) 9 way null modem cable. This is a proper 9 way cable with sealed/moulded connectors at each end so it hasn't been altered in any way. The null modem cable has 9 way sockets at each end and this plugs direct into a cheapo USB to serial connector that uses the Prolific chipset.

I'll have a look later to see if I can use it with and without the Prolific adaptor with an old laptop. I don't think the USB converter will work on this Win7 machine.

You obviously have to set it all up correctly for baud rate and I think you can check the settings by pressing the [2nd] then [Rate] buttons on the front panel of the Fluke 45.

There are other settings that can be changed as well (eg parity and echo) but I've never messed with these. But I think you can see these by pressing the [AUTO] key a few times. I think it is slow to respond/update when you scroll through these menus so be patient with it. If you want I can tell you what my settings are?

G0HZU_JMR · 13th Sep 2017, 8:53 pm

Note that it's worth getting this interface working because you can use the RS-232 interface to dump out the calibration data from internal memory and keep it safe as a file. I don't think you can alter the cal data without entering the formal calibration process and this requires the front panel calibration button to be pressed first.

So it's quite safe to dump out the cal data. I've done this with my meter.

WaveyDipole · 13th Sep 2017, 9:43 pm

Quote:

Originally Posted by duncanlowe

The user manual has information on setting up the port and seems to suggest its a standard RS232 interface rather than something with a custom pinout.

Indeed it does and refers to the RS232 which does suggest a regular serial pinout which, in turn, tends to lend weight to the adapter being the problem.

Quote:

Originally Posted by MrBungle

There are a lot of dodgy PL2303 converters out there which are missing level shifters and misbehave with certain serial port implementations.

If you can get it open, there should be a MAX211 or similar shifter on one side and an unmarked or prolific branded SSOP package on the other side.

If it's missing the level shifter or the package is a mysterious epoxy glue blob then the adapter is at fault for certain!

The fake ones don't work with my Uni-T UT61E showing the same symptoms. The real one does.

Thanks for the insight. Very useful to know! Unfortunately this one I pruchased recently is not a PL2303. It gets identified as "QinHeng Electronics HL-340 USB-Serial adapter". The plug is sealed tight so I can't really crack it open if I want to send it back. However, it is translucent enough to see just a single black blob - no other components - and pads either side of the PCB for the wires and D plug.

Curiously my computer reports a COM1 port although there is nothing on the plate at the rear so maybe there is a header on the board that I could use. Something for me to look into. In the meantime, I found a weird moulded cable with an unused weird cable with a female 9 pin plug and 3.5in jack plug at the other end. No idea what it was for, but the plug was easily removed from the moulded plastic. I was then able to transform a male-to-female cable into a female-to-female one and connected my original genuine PL2303 adapter. The result:

*IDN?
FLUKE, 45, 5295159, 1.7 D1.0
=>

Yey!

So the QinHeapOfJunk was the problem. It is obviously not fit for purpose and will be returned.

WaveyDipole · 13th Sep 2017, 10:33 pm

G0HZU_JMR, just seen your comments which apprear to have crossed with mine. Thanks for the detail. The Prolific adapter does seem to work and you are quite correct - it does require a null-modem configuration. I found that it only requires only 3 wires - GND plus the TX and RX crossed over. The DTR can be left disconnected. The settings are straightforward - 9600 baud, no parity and echo on. The Fluke does not give the option for data bits and stop bits but just assumes 8 and 1.

I need to have a look into what the process is for dumping the CAL data which would be very useful to do. I have currently tested on Linux only, but will test on Win 7/10 tomorrow.

MrBungle · 13th Sep 2017, 10:37 pm

Good stuff. Glad to see it working.

I have a cable like that floating around. It was used for connecting my TI-86 calculator to a PC if I remember correctly.

G0HZU_JMR · 13th Sep 2017, 10:45 pm

I dumped out the cal data a few years ago and I think the command is as per the service manual. I think you just send:

CALCONST? xx

where xx is a number from 1 to something like 33. It should reply each time with the cal constant for xx and somewhere in the service manual (table 5-9?) there is a table that tells you what each xx constant is for. But I can't be certain because it was so long ago. I did it all with VB and dumped the results to a file.

If/when my Fluke 45 goes badly out of cal I intend to hack the firmware (temporarily) to allow write access to the memory that holds this data. This would allow me to tweak individual cal constants without needing all the formal/automated calibration hardware/software that is required with the factory calibration routine.

WaveyDipole · 4th Oct 2017, 11:46 am

Jeremy, thanks for this information. I have finally got around to having a look at the service manual and dumping the cal constant data. You appear to have a good memory as it was indeed table 5.9 that holds the information (also duplicated in table 4.9). I now have a dump of the constant data and all is within spec. It seems a shame that there is no command to simply re-instate the previously saved value in the even of the cal data being lost for some reason and that some form of hacking it required to achieve this - or else a costly calibration by an authorised dealer.

13th Sep 2017, 5:40 pm	#1
WaveyDipole Nonode Join Date: Mar 2013 Location: Leicester, UK Posts: 2,532	Fluke 45 Serial Port Does anyone have information on the pinout of the serial port in the Fluke 45? The service manual does not seem to mention this although it goes into detail as to how the signal goes from the processor to reach the port and vice versa. I have tried to connect mine to the PC but cannot get it working. I have tried a straight cable as well as a null-modem one and neither works. If I plug my USB-to-serial adapter in directly, then all segments come on after a couple of seconds (like at startup) and stay on until I unplug the cable. I don't have a computer with a hardware serial port on it. My older PL2303 converter seems to work with everything I throw at it, but the Fluke port has the opposite gender so I have had to buy another converter with a female plug to connect with it. Since I have nothing else to connect it with to test it, I have the GND, TX and RX pins with a logic analyzer and it is transmitting. Still, I have no way to be sure whether the converter is at fault or a custom pinout has been used on the Fluke. Any information would be of great help. Thanks. Last edited by WaveyDipole; 13th Sep 2017 at 5:49 pm.

13th Sep 2017, 7:00 pm	#2
duncanlowe Nonode Join Date: Jan 2016 Location: Stafford, Staffs. UK. Posts: 2,531	Re: Fluke 45 Serial Port The user manual has information on setting up the port and seems to suggest its a standard RS232 interface rather than something with a custom pinout.

13th Sep 2017, 7:17 pm	#3
MrBungle Dekatron Join Date: Jun 2016 Location: London, UK. Posts: 3,687	Re: Fluke 45 Serial Port There are a lot of dodgy PL2303 converters out there which are missing level shifters and misbehave with certain serial port implementations. If you can get it open, there should be a MAX211 or similar shifter on one side and an unmarked or prolific branded SSOP package on the other side. If it's missing the level shifter or the package is a mysterious epoxy glue blob then the adapter is at fault for certain! The fake ones don't work with my Uni-T UT61E showing the same symptoms. The real one does.

13th Sep 2017, 7:24 pm	#4
buggies Heptode Join Date: Jan 2009 Location: West Lothian, UK. Posts: 761	Re: Fluke 45 Serial Port Hi, According to the operating manual (I don't possess that DVM), the pinout is marked on the back panel just above the connector. The drawing shows:- Pin 2 RX Pin 3 TX Pin 4 DTR Pin 5 GND The manual gives the method to detect if enabled on page 5-2. The serial defaults are 9600, 8, N, 1. It can talk to a dumb printer or terminal so even Windows should be able to cope. __________________ George

13th Sep 2017, 8:49 pm	#5
G0HZU_JMR Dekatron Join Date: Sep 2010 Location: Cheltenham, Gloucestershire, UK. Posts: 3,077	Re: Fluke 45 Serial Port I'm fairly certain it's a null modem (swapover) cable. The manual also hints that it is as null modem cable as it says to use a cable designed to link two IBM PCs together. I've got a Fluke 45 here and I have used it a few times over RS-232 and I use it with a USB to RS-232 converter and a commercial (not homemade) 9 way null modem cable. This is a proper 9 way cable with sealed/moulded connectors at each end so it hasn't been altered in any way. The null modem cable has 9 way sockets at each end and this plugs direct into a cheapo USB to serial connector that uses the Prolific chipset. I'll have a look later to see if I can use it with and without the Prolific adaptor with an old laptop. I don't think the USB converter will work on this Win7 machine. You obviously have to set it all up correctly for baud rate and I think you can check the settings by pressing the [2nd] then [Rate] buttons on the front panel of the Fluke 45. There are other settings that can be changed as well (eg parity and echo) but I've never messed with these. But I think you can see these by pressing the [AUTO] key a few times. I think it is slow to respond/update when you scroll through these menus so be patient with it. If you want I can tell you what my settings are? __________________ Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 13th Sep 2017 at 9:06 pm.

13th Sep 2017, 8:53 pm	#6
G0HZU_JMR Dekatron Join Date: Sep 2010 Location: Cheltenham, Gloucestershire, UK. Posts: 3,077	Re: Fluke 45 Serial Port Note that it's worth getting this interface working because you can use the RS-232 interface to dump out the calibration data from internal memory and keep it safe as a file. I don't think you can alter the cal data without entering the formal calibration process and this requires the front panel calibration button to be pressed first. So it's quite safe to dump out the cal data. I've done this with my meter. __________________ Regards, Jeremy G0HZU

13th Sep 2017, 10:33 pm	#8
WaveyDipole Nonode Join Date: Mar 2013 Location: Leicester, UK Posts: 2,532	Re: Fluke 45 Serial Port G0HZU_JMR, just seen your comments which apprear to have crossed with mine. Thanks for the detail. The Prolific adapter does seem to work and you are quite correct - it does require a null-modem configuration. I found that it only requires only 3 wires - GND plus the TX and RX crossed over. The DTR can be left disconnected. The settings are straightforward - 9600 baud, no parity and echo on. The Fluke does not give the option for data bits and stop bits but just assumes 8 and 1. I need to have a look into what the process is for dumping the CAL data which would be very useful to do. I have currently tested on Linux only, but will test on Win 7/10 tomorrow.

13th Sep 2017, 10:37 pm	#9
MrBungle Dekatron Join Date: Jun 2016 Location: London, UK. Posts: 3,687	Re: Fluke 45 Serial Port Good stuff. Glad to see it working. I have a cable like that floating around. It was used for connecting my TI-86 calculator to a PC if I remember correctly.

13th Sep 2017, 10:45 pm	#10
G0HZU_JMR Dekatron Join Date: Sep 2010 Location: Cheltenham, Gloucestershire, UK. Posts: 3,077	Re: Fluke 45 Serial Port I dumped out the cal data a few years ago and I think the command is as per the service manual. I think you just send: CALCONST? xx where xx is a number from 1 to something like 33. It should reply each time with the cal constant for xx and somewhere in the service manual (table 5-9?) there is a table that tells you what each xx constant is for. But I can't be certain because it was so long ago. I did it all with VB and dumped the results to a file. If/when my Fluke 45 goes badly out of cal I intend to hack the firmware (temporarily) to allow write access to the memory that holds this data. This would allow me to tweak individual cal constants without needing all the formal/automated calibration hardware/software that is required with the factory calibration routine. __________________ Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 13th Sep 2017 at 10:52 pm.

4th Oct 2017, 11:46 am	#11
WaveyDipole Nonode Join Date: Mar 2013 Location: Leicester, UK Posts: 2,532	Re: Fluke 45 Serial Port Jeremy, thanks for this information. I have finally got around to having a look at the service manual and dumping the cal constant data. You appear to have a good memory as it was indeed table 5.9 that holds the information (also duplicated in table 4.9). I now have a dump of the constant data and all is within spec. It seems a shame that there is no command to simply re-instate the previously saved value in the even of the cal data being lost for some reason and that some form of hacking it required to achieve this - or else a costly calibration by an authorised dealer.