BarGraph Displays IGT2-203R and similar

[Richard Scales]

Does anyone have any experience in driving these panaplex-like bargraph displays?

I have some IGT2-203R and am using a well publicised circuit from here: https://stromrichter-org.translate.goog/attachment.php?aid=4116&_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp

I have made various changes, I'm using an Arduino Nano for the controller, an HV module for the 245V and I changed out the R2/D2 combination for an LR8 HV regulator as I found that R2 was getting a lot warmer that I would like.

Now I am looking at the anode supply which is controlled by R8+R18 (I am only using one channel so don't have R7/R17.

It seems that when  Q8 is off, the Anode is connected to the 245 supply via 36K for current limiting purposes - that's all good. When Q8 is on, then the anode is fed from what is now the voltage divider combination of R8+R18 (98V) - also good.

However, that combination of 24K+36k is drawing about  (245/60)mA - 4mA - the power being almost 1W - so they get a little warm. I have over specified and used 5W parts. 

What I have works just fine, and, in fact I even doubled up on the resistors to share the load even further  but I wonder if there is perhaps a better way which might not generate quite so much heat?

It might seem overkill but I could make 98V using another LR8 and then use high side drivers to 'switch' between the two (like a couple of Opto Couplers or a bunch of 42/92 transistors).

... or am I being over cautious?

- Richard

[Dekatron42]

There are few professional designs in the audio business and in the ventilation flow control business who completely skip that extra 68V supply - I haven't checked the data-sheet what it says but they don't bother with that extra voltage.  Some designs use a SN75468/9 IC (75469 in case of CMOS and 75468 if TTL is used)  instead of transistors and no pull-up resistors, just hook the 68V signal to the common pin on the SN75468/9.

For all of the designs I've seen there is not one that uses a 1M resistor on the keep alive cathode, only on the anode, they all just ground the keep alive cathode pin directly.

Some designs uses a voltage between 50-80V for the 68V voltage. Some designs uses a voltage doubler/tripler to supply all voltages from a low voltage AC transformer, so maybe use a switcher with say 62.5V DC output and the quadruple that to 250V DC if there is the possibility of hooking into the circuit before rectification/smoothing?

/Martin

[leo oel]

IGT1-203 Self Scan bar graph display 

https://dotdisplay.blogspot.com/2017/08/igt1-203-self-scan-bar-graph-display.html

пятница, 31 октября 2025 г. в 12:27:23 UTC+2, Dekatron42:

[Dekatron42]

I've dug through my documents and I've been thinking about that 1M resistor in the cathode keep alive electrode and its effect on the driving of these tubes and I think it is an error to have it there as, even though it will only flow a very low current through it, it will affect the common ground point for all of the driving circuits and that will affect the level depending on how many bars that are lit up. I've never seen a circuit that incorporates that 1M resistor in the keep alive electrode for the cathode and thinking about it it looks incorrect.

Please correct me if I am wrong but that is my opinion on how these tubes should be driven.

/Martin

[Richard Scales]

Hello and thank you for your pointers - right now, the display seems to be running evenly and looks good - no significant heat is being generated in the electronics though presence of a bunch of 5W resistors seems somehow out of place when the display should only be using mA !

- Richard

[Dekatron42]

The instructions from The Burroughs kit for the BG12205 is in this thread (I cleaned the original up a bit and it is in the last post) where they use 1W resistors.

https://groups.google.com/g/neonixie-l/c/9QE680EzaTc/m/_LUcA8coBAAJ

/Martin

[Richard Scales]

I checked that schematic and as you say 1W resistors performing in the exact same way that the latter design does. I'll forge ahead with the current plan - I am using 2 x 5W parts in parallel (to get the required resistance whilst sharing the load) which get only very slightly warm!

- Richard