• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

DM160 Magic Eye help

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi, I am hoping to add the dm160 as an indicator only eye to a micro amp I have built. I have some help in getting the right voltages for the filament and plate I am a bit stuck on the grid. The data sheet describes the grid as requiring -3v (turns fully off) and the 0 volt for fully on.

So my thoughts were to first of all create a -3v supply to the grid which the has it in an off state. Then tap into my ac audio signal in the preamp, rectify it to +only, then using an opamp try and get an output of 0v - +3v.

Finally connecting this opamp out to the grid resistor would then allow the audio to drive the grid from -3 to 0v. Is any of this right?

Thanks for any input.
 
The DM160 was a logic state indicator for early digital computers, as such it is only an on/off kind off display, not a moving shadow like the more common tuning eyes or level meter tubes. If you have DM160s, always good to use them, but they aren't particularly pretty eye-candy. All the EM8* or UM8* series are prettier, although physically BIG! in comparison, and need abouy 3x the anode voltage
 
wow Arnulf, I'm amazed and chuffed to bits that any of it was right let alone all of it.:eek: Can I just check.......the tube is directly heated requiring 1 volt, so I'll have +1v on the positive side and 0v on the negative side. it the 3v swing I need in relation to the +1v or 0v.

If I tap into the post/or pre preamp signal is there a best method for creating a tap? As the amp is only 0.3watt I presume I need to ensure I dont take too much signal.

The dm160 is v small and will go nicely into my micro amp, I'm just on a learning curve and this seems hard enough. I of course dream of using the beautiful eyes available on my bigger amps. I had an old selmer with the small magic eye for the tremolo and had a matching (homemade I presume) light box made from many larger eyes. Of course I was too young to know what a fantastic thing it was, I just remember being hypnotised by it.

the dm160's dont have shadow effects but they do at least glow at different levels dependent upon input into the grid. Check this funky little put together.

DM160 frequency indicator video by merlinblencowe - Photobucket

Thanks.

Woz
 
Hi!

The DM160 has a 1.0V 25mA filament and is designed to operate off transistor circuits with a voltage change of -5 to -0.5V with a 30V banode supply. As other Members have posted it does not have a variable-length display!!!

The DM70/DM71 has an exclamation-point display which varies in length with the grid voltage, being a maximum at -0.5V and extinguishes at -10V with 90V anode supply. The DM70 has long wire leads whilst the DM71 has cropped leads designed to fit a subminature holder, the 1950s Elpico Geloso Trape Recorder used the cropped lead version!!! Other equivalents 1M1, 1M3, Y25.

Lead connections:- 1 - g, 4 f-, 5 f+, 8 - a.

For Mains use earth Pin 5 and operate the filament from 6V with a 220 ohm resistor.

You can get octal-base eyes still and these are easier to fit to panels than more modern types, but most need 200 - 250V if you're going to see them properly and unless you can find a good NOS or used one, the modern manufactured replacements tend to have very poor phosphors compared to those of the genuine valve era!!!
 
wow Arnulf, I'm amazed and chuffed to bits that any of it was right let alone all of it.:eek:

Check your original post again; you wondered if this would work:

- datasheet says negative voltage (-3V) for off, 0V for full on, I'll take your word for it since you read the datasheet
- create some sort of -3V to 0V signal source from your AC source (this is actually two consecutive paragraphs of your post)
- create this signal from audio via rectification (= the only way to do it)
- connect this input to tube grid (this is the point where "grid leak" or "grid bias" resistor connects to as well, again the only way to do it)

So yes, correct on all points you posted there.

Can I just check.......the tube is directly heated requiring 1 volt, so I'll have +1v on the positive side and 0v on the negative side. it the 3v swing I need in relation to the +1v or 0v.

In relation to +1V in this case, but trust me, datasheet provides suggestions, you'll need to tweak this value in real circuit anyhow to getr the best possible response (easiest way to do it is by changing cathode resistor). You'll also have to tweak the input voltage swing, you might need a voltage doubler afterall, or a real amplifier with opamp.

If I tap into the post/or pre preamp signal is there a best method for creating a tap? As the amp is only 0.3watt I presume I need to ensure I dont take too much signal.

Signal swing is measured in volts rather than watts. You need to determine signal swing in volts. Normal preamps output line level voltage (1-2V RMS which is almost 1.5V-3.0V once full-wave rectified without voltage doubler).

"Watt" information is useless because you don't know whether the maximum possible signal swing is 0.3V at 1A or 300V at 1 mA which both mean 0.3W.

the dm160's dont have shadow effects but they do at least glow at different levels dependent upon input into the grid. Check this funky little put together.

You could use a neon indicator (such as the venerable NE-2) in place of tube for that. Neon lamps don't require extra heating and they also display varying intensity levels depending on current across them.

For what it's worth, there are also neon lamps which fire up to a different level (physically) depending on current across them, so they work as bar indicators, rivalling magic eye tubes.
 
Hi!

I meant to add this addendum to my post (there's no "edit" button on the forum!) - In my book "Circuits for Digital Equipment" it shows the DM160 with a centre-tapped 0.5 - 0 - 0.5 V filament transformer, if you'd like to use one of the DM160 subminature indicators, it dosen't particularly matter which side of the filament your source of control voltage is referred to as these are variable-brightness indicators only!!!

However if you want to be sure and you good at tiddly transformer-making, then I'd knock up a small h.f. 5W oscillator (TIP41C etc) with a 0.5-0-0.5V floating filament supply and a 30V 10mA anode secondary. It would need to be fully screened tho!!!
 
Last edited:
Have you a photo of the amp, so we can see what style of eqipmemt the indicator is going into?

How many DM160s do you have?
A really unusual display would be a bargraph using 10 or more DM150s, driven from an LM3916 IC. But I guess you don't want to get too complicated when starting out.

The bigger eye tubes (like the 6E5, EM4, EM80, EM87) are not difficult to use and need no complicated circuitry, so don't be scared to try them. The downside is they want about 250V on the anode & target, and about 300mA heater current. The ones with phosphor on the glass (EM87, EM84), rather than on the target electrode seem to be brighter, and last longer.
 
Hi,
I once maintained a 1960s computer that used DM160s as the console display lights. Now I am going to fire up all of those DM160s using a random number generator to provide the appearance that the m/c is working. To achieve that objective I first came up against the -3v and the positive voltages required to turn off the DM160 and then to turn it on once the output is required. I think I have solved to problem but have not tested it with the Random Number generator yet.

Yes you are correct the Filaments require 1v AC at 0.03 amps and the Anode requires 50 volts. The Anode current max is .585mA. In the course of my experiments/Research I discovered that -3 volts on the grid are REQUIRED to turn the tube/valve off. The problem I have is that I need like you the -3 to turn it off and a positive voltage (above 0v) to get the tube to turn on with reasonable brightness, still within the max .585 mA anode current. As I say I think I have solved that one. Like you I had thought of an OP amp but the need for -3 was a stumbling block. I will be interested to read your conclusions

Good Luck.
Bruce McM
 
Hmmm.... I just counted 25 DM160 tubes in my junk box.

Damn! Another distraction form the project currently in hand. Using these as a bar-graph style VU indicator is a cool idea though. I'll have to give it some thought.

(I'm currently playing with some Russian IN-9 / IN-13 linear display tubes. They're fun too.)
 
To address the OP's question regarding supplies....

The "easy" way is to forget negative supplies - voltages are only relative, anyway. I would look at biasing the filament supply at or near +5 volts. One could even consider a DC filament supply fed from a 5 volt regulator. The filaments would be in parallel connected to the regulator and then a single resistor to ground to drop the remaining 4 volts (at appropriate current). That does put a DC bias along the length of the filament, but with a reasonable HT and grid drive voltage, it will have little visible effect.

If using one of the LM391x range of meter ICs where a low level output is intended to light an indicator, then a simple logic inverter (also powered from the 5 volt rail) could couple each output from the LM391x to the grid of the corresponding indicator tube. The LM391x can also be fed from the same 5 volt supply if it's convenient. Since the outputs on the LM391x chips are open-collector, they would need some weak pull-up resistors to +5v (e.g. 4.7k) to provide a clear high-level signal to the inverters.

The last bit is the HT. For the DM160 this is not that critical. They will light with quite low voltages, but a low current supply in the range 20 - 30 volts would be fine, with a simple current limiting resistor connecting each anode to the HT supply rail. Try something like a 47k for each tube and adjust up or down from there (aiming for around 0.5mA per tube when on).

Hope this helps / makes sense.
 
The DM160 is such a crude indicator that I'm guessing you're only using it for decoration, not for indicating something very precisely. In which case you can just feed the audio straight to the grid with via a cap and grid-leak resistor; that will be enough to get it dance along with the audio without the need for a negative supply (assuming the audio reaches a volt or two in amplitude).

Here's a frequency analyser thing I did for a competition, using DM160s. I used a simple ring-of-two current source for the heaters (which are in series), which is much easier than trying to get an exact voltage. These are not precision devices, so they're dead easy to use if you just want a light show.
DM160 frequency indicator video by merlinblencowe - Photobucket

After I made that video I found that lowering the anode voltage to about 40V gave me a better range between bright and dark, so the voltages you use a far from critical. All that really matters is that you don't let too much current through the heater or you'll burn it out!
 
Last edited:
Love these little DM160s. I have plans one day to make a battery powered hybrid amp using a couple of them.

These guys run so little current and a low enough voltage it would be possible to ground the grid to the cathode of the DM160 and drive the anode with an opamp directly. Wouldn't have to worry negative voltages at all. Setup the opamp on a 30V single supply and use the 'virtual ground' bias to tune in your idle glow, if any. If you feed it a non-rectified input it will just clip against the bottom rail but it doesn't matter because you'll never see it.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.