Nixie Clock - Direct Drive HV5530

423 views
Skip to first unread message

Max DN

unread,
Aug 13, 2022, 6:13:24 AM8/13/22
to neonixie-l
Hello,

I'm starting to design my first 6 digit nixie clock and I'd like to use direct drive of the tubes (my power supply can take 6 nixies without multiplexing).

So far I have built my Nixie watch, coding in ASM, worked on extreme power saving etc.

I could multiplex the nixies as I did in this project https://fb.watch/eTb69SnxEP/ , but since I have a good power supply I wanted to build a clock with direct drive of the tubes.

I prefer to code in ASM (don't ask why :)  ), I could use a 100pin PIC18 to drive each digit individually through MMBTA92 & A42 transistors. However, if I want to use a PIC with a lower pin count (say 40pins), I may use a driver such as HV5530.

Would I still get same brightness as a genuine direct drive as when using A92&A42 transistors? I suppose the switching that happens using CLK/SDA on the HV5530 is not too different than a good multiplexing. I may be wrong.

And if I were do code ASM to use HV5530, does anyone have sample ASM code? I managed to write myself the code to read/write time from an RTC using CLK/SDA so I suppose it's very similar but I cannot figure it out from the datasheet of HV5530.

Any help is much appreciated.

Max

SWISSNIXIE - Jonathan F.

unread,
Aug 13, 2022, 9:32:31 AM8/13/22
to neonixie-l
Hi Max,
The HV5530 and other HV-Series IC's are "Serial to Paralell" converters which allow to be control 32-pins (or even more with multiple ic's) from single clock and data lines. They are basically normal serial to parallel converters with high voltage transistors attached.

To operate these you will need a clock source (for easy way a gpio that goes high/low) and a data source, which means either a pin that is high or low at the time of clock. After you've sent the required amount of data (number of outputs on the IC) the first data will transfer to "DATA OUT" pin and into the next IC (if there is one).

For a 6 digit clock you need 2 pcs of HV553

I cannot provice ASM code for PIC, but here is a low level code of ATMega:

#define thePort PORTD
#define DATA PD5
#define CLK PD7

PORTC |= _BV(PC1);  //Set OE/LE Signal LOW

 for (int i = 0; i < 32; i++) {
    thePort &= ~_BV(DATA);  //Data LOW
    if ( bitRead(val_one, i) == 1) {
      thePort |= _BV(DATA);  //Data HIGH
    }
    thePort |= _BV(CLK);  //CLK HIGH
    thePort &= ~_BV(CLK); //CLK LOW
  }
PORTC &= ~_BV(PC1);  //Set OE/LE Signal HIGH

The general operation of the 5530 is as following.

1.) Drive Data pin and Clock pin LOW
2.) Before starting to send data, drive LE pin LOW
3.) For each bit you need to do a clock signal and data pin, for this:
3.1) Drive CLOCK High
3.2) Drive DATA high or low, depending if you want to turn the digit on or off
3.3) Drive CLOCK Low
3.4) Drive DATA Low
4.) This needs to be done for all Inputs of an ic, if you have two ic's chained you need to to it two times.
5.) After all Data has been sent, drive LE pin HIGH, Outputs will now be set to the data you've just send

gregebert

unread,
Aug 13, 2022, 10:46:58 AM8/13/22
to neonixie-l
Be sure to use a level translator to drive the 5530-type devices at the proper signal voltages (12 VDC). Some people have driven them at 5V, which is out-of-spec and not guaranteed to work.  I use these drivers in some of my clocks and they work very well. 

Pardon my rant below, but I've been designing IC's for a living since the 1980's and too often I see others having problems with serial devices  :

  One thing that annoys me about the HV5530 (and similar) devices is that the datasheet specs are incomplete for the serial data signals. The data-in hold time (tH) is 10nsec, but there is no minimum propagation time in the spec (tDLH, tDHL). The only have a maximum (100ns). What this means is that per the datasheet, you cannot simply cascade multiple devices. It might work if you do, but if you want to guarantee that it will always work, place a rising-edge flip-flop between cascaded devices). In order to have reliable shifting, the minimum propagation-delay of the driving device must be greater than the hold-time of the next device. If there is no minimum prop-delay spec, assume it's zero (a very safe and reasonable assumption). 

My current clock project (6-tube RZ 568m) has multiple PC boards with ribbon cables interconnecting them. Not just for the three HV5530's that drive the tubes, but also many other devices in the same serial chain for reading switches and controlling lamps. After many weeks of testing with billions of bits, there are zero bit errors.....because I made sure setup and hold times are met.


Max DN

unread,
Aug 13, 2022, 11:40:39 AM8/13/22
to neonixie-l
Thank you both.

Say that I want to drive high HVOUT12 on pin2 (that is binary b'00000010') of the HV5530... and reading your previous messages, do I simply...:

drive pin28 (CLOCK) low
drive pin31 (LE) low

drive pin28 (CLOCK) high
drive pin32 (DATA Input) low ; send 0 for bit 0 of b'00000010'
drive pin28 (CLOCK) low
drive pin28 (CLOCK) high
drive pin32 (DATA Input) high ; send 1 for bit 1 of b'00000010'
drive pin28 (CLOCK) low
drive pin28 (CLOCK) high
drive pin32 (DATA Input) low ; send 0 for bit 2 of b'00000010'
drive pin28 (CLOCK) low etc
etc...
drive pin31 (LE) high.

Of course, I'd need to do it do it within a certain time frame (nS, I suppose)

Then I guess I would need a level shifter as while the input voltage is 12V, the PIC is at 3.3V (LF series). Argh. 

I'm starting to think whether that is really a better approach than using an 80pin or 100pin PIC with HV transistors to drive the cathodes of the Nixies (pic18f46k22 would do for instance). It would increase component count and cost of PIC etc, but it's not a commercial project, I'm only making 3 clocks in total.

Indeed, I'm building a clock with Dalibor's RZ568M (beautiful!).

Bill Stanley

unread,
Aug 13, 2022, 12:32:46 PM8/13/22
to neonixie-l
I use that part for my NIXIE, Panaplex and LED displays and I also write in assembly (8051, not PIC)
For my code, I send the serial data for NEXT time to display in the time after a 1 second update. On the 1 second "hack"
I update the display by doing the serial to parallel pulse.

Here is the code I use:

;************************************************************************
;*                                    *
;*    Serial clock the data in the raw buffer to the H/W drivers.     *
;*    Note: this routine does not strobe the data to the display.    *
;*                                    *
;*    On Entry: None.                            *
;*    On Exit:  Display H/W drivers updated from ram.            *
;*                                    *
;************************************************************************

WRT_NIXIE_HW:    MOV    R0,#DISP_RAW_TH        ;point at start of raw display ram
                                 MOV    R1,#12d            ;12 bytes of data to send
RAW_BYTE_LP:     MOV    R2,#5            ;5 bits per byte
                                MOVX    A,@R0            ;get a 5 bit value
                                INC    R0            ;bump the pointer

RAW_BIT_LP:        RRC    A            ;bit 0 into carry
                               MOV    DISP_DAT,C        ;put the bit into the I/O port
                               CALL    HV_BIT_DELAY        ;delay
                               SETB    DISP_CLK        ;clock high
                               CALL    HV_BIT_DELAY        ;delay
                               CLR    DISP_CLK        ;clock low
                               DJNZ    R2,RAW_BIT_LP        ;loop 5 bits
                               DJNZ    R1,RAW_BYTE_LP        ;loop 12 bytes
                               RET                ;and return


;************************************************************************
;*                                    *
;*    Update the NIXIE readouts with the data in the H/W buffers.    *
;*                                    *
;*    On Entry: None.                            *
;*    On Exit:  NIXIE displays show updated values.            *
;*                                    *
;************************************************************************


UPDATE_DISP:    CALL    HV_BIT_DELAY        ;delay
                              SETB    DISP_STRB        ;strobe the data
                              CALL    HV_BIT_DELAY        ;delay
                              CLR    DISP_STRB        ;un-strobe NIXIE drivers
                              RET

Max Di Noi

unread,
Aug 13, 2022, 1:04:21 PM8/13/22
to neoni...@googlegroups.com
Thanks.

In terms of nixie brightness, is there any advantage in using serial to parallel converter+shift register vs multiplexing using high voltage transistors on anodes and cathodes of the nixies? Other than saving component and uC pins, of course. 
--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/c2e49484-d56e-4545-b635-942952d2c1f4n%40googlegroups.com.

SWISSNIXIE - Jonathan F.

unread,
Aug 13, 2022, 3:51:54 PM8/13/22
to neonixie-l
The time to shift out the bits is not critical, the clock rate is as fast as you implement it, there is only a minimum clock time, but no maximum one.
Brightness should be the same as putting the tube pin directly to ground, the internal resistance of the mosfet in the HV5530 is in the Ohms range, while your anode resistor will probably will be a few kiloohms, so no visible issue.

About voltage:
The HV5530 is specified for 10.8 to 13.2V, but alot of people seem to use it with 5V logic and it works. If you want do operate ith with 12V logic, you need a 12V source and a level shifter. HEF4104 could work

Bill Stanley

unread,
Aug 13, 2022, 4:00:45 PM8/13/22
to neonixie-l
Using the high voltage serial to parallel part (I use the HV5812PJ-G ) is direct drive in that the NIXIE digits are driven with a stable readout until the digit(s) change
to update the time.
The multiplexed drive uses (typically) open collector NPN transistors connected to the 10 cathodes tied together for all tubes i.e. all 0 cathodes, all 1 cathodes etc.
The anodes with current limiters are separately driven to the high voltage rail.
The driver turns on the desired cathode digit and then turns on the anode of 1 tube. the tube in then blanked and the cathode of the next tube is set up
and that anode is activated. Persistence of vision makes the viewer think all 6 tubes are active at the same time (if the scan rate is high enough)
Since for a typical 6 digit clock, each tube is on for only 1/6 of the time, the current is raised above the normal value for direct drive to get the perceived brightness
closer to the direct drive.

To use a muxed display you to take care not to exceed the current limit for the tube and pay attention to the tube turn-on and turn off time to select the mux rate.
I have read some comments on this board RE the effects of muxing (and brightness) on a tube life. Someone with a lot more NIXIE-fu than I have could add to this.

For a 6 tube clock you would need 10 cathode drivers and control bits and 6 anode drivers and control bits for 16 GPIO bits.
With the HV5812PJ-G type devices, you need only 3 bits irrespective of the tube count. The down side it that I assume the serial HV drivers may be hard (impossible) to find right now.

Moses

unread,
Aug 13, 2022, 4:08:19 PM8/13/22
to neonixie-l
I believe the shift register speed is rated to 8mhz in these series of HV shift registers. But its not usually a problem since most people will probably not drive them that fast.

I did run a similar chip,the HV5122 at 5v with a 3.3v logic input (!) on a prototype of my last clock. It worked OK if the speed was kept low, around 100-200khz or so, which was still plenty fast to update tubes. Letting my microcontroller free-run the shift register into the mhz range broke this setup. Other then an occasional glitch I didn't see any other issues if the speed was kept slow. It was going into production, so I eventually updated the design to power the drivers with 12v and logic level shifters.

Regards,
-Moses

Max Di Noi

unread,
Aug 13, 2022, 4:21:52 PM8/13/22
to neoni...@googlegroups.com
Great. I found a nice page that confirms the sequence of CLK/Data, I think it's in line with my understanding of your suggestions. http://reboots.g-cipher.net/time/

So, the code for a serial to parallel convertershould be easy to implement.

Yes, I'm aware that some people use the hv5530 at 5V but I'd like to use it in spec. I'll use a Hef4104 as you suggest.

So I supply put it in between the PIC and the Hv5530, using 2 pins of the hef4104 on the input side and 2 pins to the CLK/Data of the hv5530? I've never used it before but I understand it should be simple. 

Max Di Noi

unread,
Aug 13, 2022, 4:23:22 PM8/13/22
to neoni...@googlegroups.com
Thanks. So, that confirms that multiplexing reduces brightness. Which makes sense now that I start to understand how the HV5530 works.

Max Di Noi

unread,
Aug 13, 2022, 4:35:13 PM8/13/22
to neoni...@googlegroups.com
Jonathan,

I just found this, it's excellent. 


From this post:

If the code is open source, I'll just use that otherwise I'll adapt mine to use hv5530 and hef4104. 

Awesome! 


-------- Original message --------
From: "SWISSNIXIE - Jonathan F." <jfre...@gmail.com>

SWISSNIXIE - Jonathan F.

unread,
Aug 13, 2022, 5:16:06 PM8/13/22
to neonixie-l
Yes, the code is open source :)

You can download it at the bottom of the page


I programmed it in Arduino with the atmeag328, but it should be a good start to look at the function "setOutputs" in the code, it contains the actual shifting process.
Also it's easier to layout the outputs of the IC and cathode-pins as "numbered" with the output number unless you are very experienced with the code.  This allows bitwise operations for creating the variable containing the value.
For example, if your Cathode 0 is connected to output 1, and cathode 9 to pin 10, you can use a function to shit as a single bit according to the current digit number. If your digit is 8, you can do " 0x01<<8". If your cathodes are randomly connected (to make an easy layout) you will need to implement more code.

Max Di Noi

unread,
Aug 13, 2022, 5:28:27 PM8/13/22
to neoni...@googlegroups.com
You are a legend! I've looked at the schematics, very neat.

I'm traveling on business tonight, back next weekend. Then I'll design my own board for 4 beautiful Rz568m (present for my parents on Christmas). 

Glad I now understand how the HV5530 etc work. 

Looking forward to playing around with the code. 

I'll definitely include RTC, GPS receiver and IR motion sensor (I finally designed and tested a circuit using the cheap Ds203 ir sensor. I'm trying to design a microwave sensor but only with minor success so far, the antenna is very challenging to design without proper training and measurement tools). 

Thanks everyone for jumping in with so many suggestions. 

I'll send a pic of my clock when ready. 

Richard Scales

unread,
Aug 14, 2022, 12:41:35 AM8/14/22
to neonixie-l
I've used HV5522 and HV5622 a lot without any issue - I always use a CD40109B level shifter to convert from 3v3V or 5V to 12V (I know many do not).

In Arduino Land there is a function called ShiftOut which does all the required clocking - it's for 8 bits as standard but can easily be re-worked for however many bits you want.

I have one clock with 4 x HV5522 in series for which I made a 128 bit ShiftOut function.


It's an interesting page with scope traces to help explain.

Ultimately I moved over to using SPI transfers which are driven purely by the processor but the theory is all very similar.

- Richard

Michail Wilson

unread,
Aug 14, 2022, 4:35:32 AM8/14/22
to neoni...@googlegroups.com

Good that you mentioned ShiftOut.   On that note…

 

I use an ESP-8266 for most of my projects (specifically the WeMos Mini D1).

Limited in PINs, but I want to control many steppers, so I use the mentioned ShiftOut.

 

Basically, using 595s (4 of them), I control 8 stepper motors.   A 595 having 8 outputs and the steppers need 4, I get to control 2 per chip.

 

In the reverse, I have 8 switches (end points/stops) for the 8 steppers.

 

All happily controlled by a single WeMos.

 

Of course with 8 steppers * 4 wires each…. It’s a jumbled mess in prototyping.

 

Michail

 

From: neoni...@googlegroups.com <neoni...@googlegroups.com> On Behalf Of Richard Scales
Sent: Saturday, August 13, 2022 9:42 PM
To: neonixie-l <neoni...@googlegroups.com>
Subject: Re: [neonixie-l] Re: Nixie Clock - Direct Drive HV5530

 

I've used HV5522 and HV5622 a lot without any issue - I always use a CD40109B level shifter to convert from 3v3V or 5V to 12V (I know many do not).

gregebert

unread,
Aug 14, 2022, 11:00:50 AM8/14/22
to neonixie-l
One other thing to add....It's good practice to loop-back the end of the serial chain to your controller so you can run diagnostics on the serial chain.

On my projects, I run the clock and data signals back to the Raspberry Pi or FPGA so I can verify the integrity of the chain. If the clock doesn't wiggle, the chain is definitely broken. The other reason is that looping-back the clock allows you to drive clock and data signals in the opposite direction, which improves the timing margin for hold-time. This has been common practice in chip design for decades. Reading back the data and comparing against a delayed version of what was sent will tell you if there are any shifting problems. You can also develop code to measure the length of the chain, set or sample specific bits, and  speed tests (FPGA; RasPi is too slow).

David Pye

unread,
Aug 15, 2022, 10:09:33 AM8/15/22
to neoni...@googlegroups.com
Given the price of these custom drivers, I wonder whether some cheap arm microcontrollers with serial interface and mpsa42 type transistors would be a decent replacement.  Especially if getting the boards preassembled by jlcpcb or similar, where parts count is less of an issue than by-hand assembly.  Something like the gigadevices ST clone. Gd32e130 is about 2usd at worst.

Can easily speak to them with 3v3 as well which is nice.

David

David Forbes

unread,
Aug 15, 2022, 10:46:03 AM8/15/22
to NeoNixie
The most sensible low-cost direct drive arrangement in terms of parts count and board layout is SOIC 74HC595 shift registers and SN75468 HV driver arrays. 
There's a Chinese Zirrfa 6 digit clock board that uses this approach. 

Mark Moulding

unread,
Aug 15, 2022, 1:56:56 PM8/15/22
to neonixie-l
I almost invariably use daisy-chained 74HC595s in my clocks.  I use Numitrons rather than Nixies, which can be direct-driven with 5 volts, and an 8051 variant (AT89C4051) because I have lots of them and a tool chain I'm comfortable with; they're programmed in C.  To drive the clock, data, and load lines, I just directly bit-bang the micro outputs in code, and the '595s are fast enough that I don't need to worry about timing - I just go as fast as I can.  I never bothered to do a timing analysis because I figured I wasn't at all close to the timing limits of the parts.  Another pin from the microprocessor is used to apply a PWM signal to the output enable of the '595s to provide dimming of the display.

Also, the '595s have a high output drive capability - they're rated at 35 mA, so driving the individual segments of the Numitrons (~20 mA each) is no problem.  I have many thousands of hours on several different clocks, over more than a decade, with zero failures of either the drivers or the tubes.  (Occasionally, I'll lose a power supply...)
~~
Mark Moulding

gregebert

unread,
Aug 16, 2022, 4:44:42 PM8/16/22
to neonixie-l
If bit-banging is done serially, you can avoid most timing problems by-design. However, if the controller has an 8-bit parallel-output port, and data is changed  on the same cycle as your sampling clock-edge (usually rising-edge, but HV5530 and similar use falling-edge) you create a race condition that can lead to an error. Some controllers are so fast, even doing writes sequentially isn't a guarantee, as I describe below.

A few years ago, I still had a rare failure while bit-banging between a RasPi and an FPGA. After a few hundred megabytes, I got an error on 1 bit. It was caused by a race-condition with the clock signal. Adding a short instruction between writes solved the problem, and there have been zero errors since then (roughly 4 years and a few terabytes or more of register traffic).

ZY

unread,
Aug 17, 2022, 11:38:46 AM8/17/22
to neonixie-l
Don't numitrons have pretty high shoot-through current? I'd expect 100mA of current going into the IC pins at each off-on transition. Which would probably be fine since it's only for a few milliseconds and as you mentioned it's been fine for years, but it's technically out of spec.

Christian Riise Wagner

unread,
Aug 17, 2022, 2:12:03 PM8/17/22
to neonixie-l
Even running more than 3 segments @20mA is actually out of spec as the maximum ground and source current for the 74HC595 is 70mA. I found that it didn't work satisfactory for an IV-19. Too high a voltage drop. Might be okay for smaller numitrons though. I've ordered some TPIC6C595 instead to try out instead. 

Regarding the software, if the microcontroller has an SPI peripheral (most do), you can use that to send out the serial data instead of bit banging. It's how I've done it on all my microcontroller-based clocks so far. Works on all the 3 different architectures I've tried. It's slightly more efficient than bit-banging. The performance increase mostly won't matter, but it does make sense when you're doing cross-fading of the digits where you might be updating the registers hundreds of times per second.

liam bartosiewicz

unread,
Aug 17, 2022, 4:29:36 PM8/17/22
to neoni...@googlegroups.com
I’ve had great experience with the TPIC6B595s, I used them to drive a whole bunch of miniature bayonet style bulbs for a clock using IEE one plane readouts, no problems with them so far. Probably a little over 120 mA per IC. 

On Aug 17, 2022, at 8:12 PM, Christian Riise Wagner <c.riis...@gmail.com> wrote:

Even running more than 3 segments @20mA is actually out of spec as the maximum ground and source current for the 74HC595 is 70mA. I found that it didn't work satisfactory for an IV-19. Too high a voltage drop. Might be okay for smaller numitrons though. I've ordered some TPIC6C595 instead to try out instead. 


Regarding the software, if the microcontroller has an SPI peripheral (most do), you can use that to send out the serial data instead of bit banging. It's how I've done it on all my microcontroller-based clocks so far. Works on all the 3 different architectures I've tried. It's slightly more efficient than bit-banging. The performance increase mostly won't matter, but it does make sense when you're doing cross-fading of the digits where you might be updating the registers hundreds of times per second.

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

Mark Moulding

unread,
Aug 18, 2022, 12:39:26 AM8/18/22
to neonixie-l
Regarding using the 74HC595s, it's true that they are technically out of spec if too many segments are on, but I believe the failure cause would be whole chip heating rather than the single output transistors, and especially with the PWM dimming going on (the segments are never on 100%, since that would be out of spec for *them* - they're actually rated for 4 volts, not 5), everything stays nice and cool.  I haven't had any problems with ground bounce screwing up the logic states.

The cold-filament current could be a real deal, and in fact happens when power is first applied to the clock.  I have a start-up routine that sequentially warms up each filament to just below visibly glowing, so that condition only exists for less than a second, and only 100 mSec per segment.  After that, the filaments are all kept warm.  Actually, the original RCA data sheets recommend doing something like that to limit the current surge of the cold filaments.  I have a couple of old clocks that don't have this algorithm, using my precious DTF-104B tubes, and it doesn't seem to have unduly affected them either, though.

Regarding the timing of the bit-banging, the particular processor I'm using avoids all of those problems, mostly by virtue of it being rather archaic.  Firstly, the port pins in an 8051 are in an area of memory that's bit-addressable, so that glitch when the the entire 8-bit port is written doesn't exist.  It's also a couple of orders of magnitude slower enough so that it won't outpace the specs of the '595s.  (I did put spacing instructions in when bit-banging some slow SPI peripherals on another project, though.)  This old, tiny processor certainly doesn't have any built-in SPI peripheral, though.

I'll have to look into the TPIC6b595s - those sound useful...
~~
Mark Moulding

Christian Riise Wagner

unread,
Aug 18, 2022, 1:40:36 PM8/18/22
to neonixie-l
I agree. It's probably very unlikely that you'll damage an output by simply pulling too much current. I guess the internal resistance is too high to cause immediate destruction of the output FET. I didn't have any problems with the logic either. Just too high voltage drop across the '595 causing too low brightness of the IV-19. Didn't check temperature, but it must've been dissipating some power. Glad to hear it worked for you though!

If your microcontroller doesn't have SPI there's no question of course. But regarding the speed, I'm running a 10MHz SPI clock on my ESP32 based clock without any issues. The datasheet even says they're good for up to 100MHz. Never tried more than 10 though as it wouldn't significantly improve transfer times. I do see though that the HV-series shift registers have a max specified speed of 8MHz. But all the SPI capable microcontrollers I have worked with has had some means of dividing the master clock to lower the clock speed. There may be exceptions though.

As for the (S)TPIC series, the '6B595 is the higher voltage rated version (50V) and comes in a larger package and costs more. For lower voltage stuff you can get away with the '6C595 which is rated at 33V.

Max Di Noi

unread,
Oct 3, 2022, 7:15:13 PM10/3/22
to neoni...@googlegroups.com
Hi Jonathan,

I have built this clock and works well. I have however decided to go for a 4 digits version instead of 6. I can manually set the time, no issues. But when I go through the parameters menu, I can see the menu number but not the parameter numbers of each menu as the code is displaying those using the second digits which I don’t have.

I can code in Assembler but I am not very good with Arduino yet. Is there an easy way to modify the code for 4 digits only?

Any help would be much appreciated.


Thank you,

Max



SWISSNIXIE - Jonathan F.

unread,
Oct 5, 2022, 1:05:31 AM10/5/22
to neonixie-l
Hi

If I understand correctly, you need to alter the code in the section "menuItemHandler"

The function calls on line 639 and 654 do set the values to the three tube groups, if you do not have the seconds column, you would have to change this this:
formatOutput(menu_item, 255, currentVal, 0, 0);  and set the currentVal to the second tube group resulting in formatOutput(menu_item, currentVal, 2550, 0); 

Max Di Noi

unread,
Oct 5, 2022, 2:10:24 PM10/5/22
to neoni...@googlegroups.com
Beautiful, works like a dream! 

I’ll go on and play around with GPS and Motion sensor features.

If you don’t mind, I’ll ping you again for anything quick but I’ll try to understand the code myself first.

You did save me a couple of nights going through the code though :)

Thanks so much.

Max


Reply all
Reply to author
Forward
0 new messages