Forum Vibe - bad lamp?

[Brossman]

Hey all,

I've purchased the most recent version of Brad's Forum-Vibe, began populating, and now having trouble.  As per RG's advice long ago, I began with the LFO section - get that running, then deal with audio...

So, AFAIK, the oscillator is working (I see voltages on Q11/Q12 pair changing - if posting these ranges would help, let me know).  Aside from that, the bulb just wont light... I'm thinking that the trimmers I have for the Offset adjustments and bulb Bias may be busted somehow...  I see voltage on either side of the bulb (~7v) but I'm not sure of this one's specs (though it did test moderately strong with a 9v battery).

Any advice here? Thanks in advance,

- Britt

[R.G.]

Details matter. If it were me, I'd
1. Use my meter set to resistance to measure the resistance of the bulb. If it's open, it won't light. If it's not open, it's most likely good.
2. Meter the resistance of the trimmers. They should be a fixed resistor from end to end, variable from wiper to ends. If yes, they're good.
3. (Re-)read "debugging: what to do when it doesn't work" and follow it. The DC voltage measurements are deadly effective in finding bad DC conditions that prevent things from working. If two points that are supposed to be connected by a copper trace are at different voltages, they are not connected. If two points that are supposed to be different voltages are the same, exact voltage, they may well be shorted and should be tested by ohmmeter. If an NPN doesn't have its emitter lowest, base about 0.5-0.6V higher, and the collector a bit higher than that, it cannot be amplifying. There are so many things that fall out of the DC conditions.

[Brossman]

Thanks RG, I'll double check that and get back.

OKAY... From "De-bugging"
1. This is Brad's (redhouse) Forum Vibe (board FV4, double-sided pcb)
2. Here is the layout of the part numbers which I will be referencing http://classicamplification.net/forumvibe/fv_mod_08.htm (though my board's layout is different, as are some of the component values).
3. I followed Brad layout as closely as possible, though the FV4 board is a *bit* different than v3 that is currently on the website (see link)
4. a. I used 2n5089 for Q2-Q9 and Q11-Q14, with Q1 as 2n930 (TO-18 package) and Q10 as NTE46, an MPSA13 sub.
    b. As per RG's "Tech. of..." page, I swapped C5, C9, C12, and C15 with 10uf caps, as I didnt have enough 1uf to populate, but had oodles of 10uf's... (plus, who doesnt like more notch depth in a phaser?)
    c. C16 omitted (as per JC's advice on his website).
    d. R31 (68k) - 100k subbed (also from JC, to lower noise...)
    e. R36 (47k) for vibrato out - 100kohm subbed.
    f. as all of these deal with the audio side, I'm really not sure if they'll be relevant to this issue (but are included for posterity).
5. PSU is a 24vDC (200mA) wall wart into independent 7815 regulators (with appropriate filtering) for audio and LFO power. Common power ground (bad idea?)
6. I have not worked with the audio yet, as I am out of jacks (and refuse to use the crap at Rat Shack)...

Now then,

1. my trimmers *seem* to be fine... check.
2. the bulb has a resistance of about 20R... check.
3. hmmm DC... I'll get this one in a minute (for now, this).

[Brossman]

Before I begin with this information, AFAIK, I have double and TRIPLE checked my tranny orientation... (over and over and over and...)

NOTE: Due to oscillations, readings are ballparked...

Q11
C =14.94
B =6.3-7.4
E =7.5-8.8

Q12
C =14.94
B =7.2-8.8
E =6.7-8.2

Q13
C =14.86-7
B =2.6-3.0
E =2.1-2.5

Q14
C =14.86-7
B =2.1-2.5
E =1.5-1.9

If you ask me, Q11's base voltage looks funny... a place to start?

EDIT::

Starting from that I began checking Q11-Q14 with the Diode function on my DMM, finding that Q13 has only one functioning diode between emitter and collector. I will try swapping this...

[Brossman]

So I stupidly replaced that transistor, not realizing until afterwards, that there is a trace from Q13E to Q14B, and Q14C to Q13C, Making my test see the one diode from q14B_C when TRYing to see between Q13 E-C... *sigh*

I was becoming a bit frustrated and so was that spot on my board, so I set it down and modded the crap outta my buddy's Russian BMP to make myself feel better.

This being said, I still have not discovered the cause of the issue.  I took a 9v battery (9.42v) and strapped it across the lamp and, voila - light.  So I havent blown the bulb... ALSO, when I apply power to the circuit, there is ~15V at the bulb. Why no light?

most people have oscillator issues, but I cannot even get the dang bulb lit...

[R.G.]

This being said, I still have not discovered the cause of the issue.  I took a 9v battery (9.42v) and strapped it across the lamp and, voila - light.  So I havent blown the bulb... ALSO, when I apply power to the circuit, there is ~15V at the bulb. Why no light? 

Is there 15V on BOTH sides of the bulb?

As Georg Ohm said, V = I*R. You have proven that the bulb has some resistance however small. So the only way it could have the same voltage at both ends is if the current through it is zero, and that would mean the transistor driver is not pulling current.

[Brossman]

Yes, RG, that would seem to be the case. I have (IIRC) the same voltage at the Collectors of Q13/14 that is on either side of the bulb (~20ohm). 

Why would it not be pulling current? Is my Bias trimmer set too high?

[R.G.]

Yes, RG, that would seem to be the case. I have (IIRC) the same voltage at the Collectors of Q13/14 that is on either side of the bulb (~20ohm). 
Why would it not be pulling current? Is my Bias trimmer set too high?

Have you removed the lamp and tested for a PCB short across the pads of the bulb? I belatedly consulted Georg Ohm myself and he said that current would be nearly zero if the PCB has  a short too. This is just after I looked at the voltages for Q13/14. 

[Brossman]

I'll have you know that by "14.86-7" I mean that it is varying between 14.86 and 14.87 on the meter.... I hope that wasn't throwing you. 

Either way, I'll do this with the bulb and get back.

[Brossman]

Okay, with the bulb removed, there is no continuity between pads (so I guess no short, yea?)

With respect to ground ( - power terminal for LFO), I get 14.93V on the side that connects directly to the + terminal, and .575-.593V oscillating on the other.  So If I understand correctly, the oscillator IS working, but I'm just not drawing enough of "something" to power the bulb, right?

As I find it might be relevant, these voltages were taken with the Bias trimmer at 500R (in series with ~150R). When I turned the trimmer down to ~250R, I read 0.559-.572V on the other pad... Stick the bulb in the holes didnt light it...

EDIT::
For fun, I decided to take a red LED I had sitting around and attempt to light it in the bulb's place... It lights! Changing speed affects it! I really dont know what to do from here....

[R.G.]

Looking back on the posts:
- the lamp's voltage stays near 15V on both terminals
- the lamp is not open, and glows from another power supply
- the PCB is not shorted across the lamp.

That leads us back to the driver transistors, and perhaps to the lamp itself.

What's the lamp type and voltage/current rating?

And right now is where you want an oscilloscope, because we know from the battery test that it glows with 9V across it. Reading from your earlier posts, the low side of the lamp goes down to about 7V. That's not quite 9V across it, only about 7.5-8. The question the scope answer is how low does the low side of the lamp really go.

Try this. Turn the depth control down to zero. This stops the oscillator from mucking things up. Measure the low side of the bulb (= collectors of the drivers) and twist the trimmer from end to end, noting the voltage the lamp/collectors hit at each end of the pot turn. Also what the power supply voltage is at each end of the pot adjustment.

Could be the power supply is collapsing, too.

[Brossman]

That leads us back to the driver transistors, and perhaps to the lamp itself.

What's the lamp type and voltage/current rating?

I suspect this... As it has been way too long for me to remember what the rating of these bulbs are, suffice to say, grain of wheat, probably 9-15V range-ish...

Reading from your earlier posts, the low side of the lamp goes down to about 7V.

That was a typo  sorry. It stays around 14.8V

With the Bias trimmer at 500R (in series with ~150R)... I get 14.93V on the side that connects directly to the + terminal, and .575-.593V oscillating on the other.
...trimmer down to ~250R, I read 0.559-.572V on the other pad...

Could be the power supply is collapsing, too.

I really don't know how two 7815's could be collapsing, but it is a possibility.  I have 220uF on the input of each, protection diodes, and 22uF for transients on the output - seems pretty solid.

At this point, I'm ready to get this puppy *working* before making all of my googly-boogly, add-on, switching-array, fun-time knobbies.  To this end, how about a nice uh... white LED if the bulbs aren't working? (BTW, tried checking both bulbs on a 9v again, and they seem to be ..fried?   )

[R.G.]

I suspect this... As it has been way too long for me to remember what the rating of these bulbs are, suffice to say, grain of wheat, probably 9-15V range-ish...

If the current/power rating was too high, the transistors could not pull down enough current to light the bulbs. A 9V battery can supply upwards of an amp or two if it's suitably shorted, so it can power a high current bulb for a little while.

That was a typo  sorry. It stays around 14.8V

OK - I was trying not to miss something. This is another vote for an incompetent driver transistor or overly demanding bulb.

I really don't know how two 7815's could be collapsing, but it is a possibility.  I have 220uF on the input of each, protection diodes, and 22uF for transients on the output - seems pretty solid.

It's more a function of the raw DC than the regulators and things around them. If the raw DC drops too low, the regulators can't hold up their output no matter how good they are. But since you caught the typo on the drop of the low side of the bulb, it's not likely that this is the issue.

If you're using a digital meter, you might well miss a short drop lower than normal voltage because digital meters take samples at certain invervals, then display that sample. Watching a varying voltage or current is one of the few things that analog meters do better than digital ones.

At this point, I'm ready to get this puppy *working* before making all of my googly-boogly, add-on, switching-array, fun-time knobbies.  To this end, how about a nice uh... white LED if the bulbs aren't working?

A white LED will work, but vibe purists insist that the time response of an incandescent bulb is necessary for some of the quirks of the original circuit to shine through. You may not care.

(BTW, tried checking both bulbs on a 9v again, and they seem to be ..fried?   )

Using your meter, are they open/high resistance? 

[Brossman]

I insist that there be a bulb as well, but I'd rather see it *work* than not at this point.

Regarding the bulb, new bulb - 21.7R

For posterity, I took DCV at either side of PSU: from a 24DCV, 200mA wall wart, I get 30.7V on pin 1 of LM7815  , and 14.9V on pin 3... again, seems solid...

RG, I'll get more info to you (re: Q13/14DCv) when I get a new battery for my analog VR meter. If need be, I do have an O-scope, albeit old (Sencore PS148A), though I'm a bit unfamiliar with its use.

[R.G.]

I insist that there be a bulb as well, but I'd rather see it *work* than not at this point.

Very engineer-y of you.    Do what works first, do the perfect thing once that's discovered.

Regarding the bulb, new bulb - 21.7R

Good reference info.

For posterity, I took DCV at either side of PSU: from a 24DCV, 200mA wall wart, I get 30.7V on pin 1 of LM7815  , and 14.9V on pin 3... again, seems solid...

That's enough headroom.

Let's see. We have issues between the bulb and the drivers. The bulb seems OK, but doesn't light. An LED lights. An LED needs only about 2-20ma. The bulb should only need maybe 25ma. Does the LED fade in and out when depth is turned up? If so, the driver is getting the LFO through the depth control.

Have you had a chance to get voltages on "15v" and the collectors of the drivers with the depth control full down and in both min and max on the trimmer?

Let's also do some thinking. That emitter resistor for Q13/14 is 150R minimum, 650R maximum. Current through the bulb can go as high as 50ma, so the voltage on that 150R resistor is as small as 0, and as large as 150*0.04 = 6V. For a 25ma current, that's 150*0.025 = 3.75V, which is easy to measure.

Try turning the depth fully down, and measure the voltage across the 150R resistor when the trimmer is min and max. That should tell us how much current is coming through the drivers for various resistance settings of the trimmer.  The forum vibe also includes a base bias trimmer as well as an emitter trimmer. What's populated there and what does that trimmer do to the voltage across the 150R ( = emitter current on the driver).

[Brossman]

Okay, I've been doing extensive work debugging on my own...  Mr. Ohm says that if my voltage is constant and I'm not getting enough current to drive my bulb, then it would be logical to lower the resistance on the trimmer, right?

I will note that I acquired new bulbs (#18, T1 3/4, 14v, .04A). I lowered the resistance to zero (0) on the trimmer giving only ~150R to the bulb, and there was a slight red color to the filament...  I replaced my 150R resistor with a 22R and put the trimmer at ~100R. Soldered in the new bulb and fired it up. I HAVE GLOW. Messing around with the bias trimmer, I was able to dial in a nice pulsing!!

**I am not sure why my lamp driver is so weak, but for the time being, this solution, albeit a bit unconventional, has worked splendidly.**

As for the audio, I wasn't getting signal through except a sputtery splatting... I was convinced the problem lay with Q1. I double checked and sure enough, had it flipped around.  I used a BC547 from a Russian Big Muff.

RG, thanks so much for all of your help!

[RedHouse]

RG, thanks so much for all of your help!

Ditto!