Late 50' tube... who guess ! (The Ondioline)

[zamproject]

Hey

You definitely should not be able to touch a wire with 140 volts on it! That will hurt! Are you sure its not some sort of capacitor arrangement? So the wire will be insulated, but press up against the other wire to change capacitance?

There is capacitive coupling involved in the system holding the sound attack/note-on trig, via the "boite d'attaque expressive" (expression attack box...another safety issue due to asbestos fibre in it, wish I will find substitute)
But for the understanding and (passive) measure I have now, the "corde" system is totally an on/off DC switch...from HT to 6BA6 G2)
There is DC capacitor discharge involved in the "percussive" network (around P switch) to aslo drive the pentode screen with keayboard dedicated switch but this is an alternative function decorrelated to the expressive box (drive grid via lower A switch assembly), which again differ to the "percussion" setup

If my math are good, we have 0.3mA here, which under 140v is 40mW.
Theoretically no issue for human body, but what if there is a fault (short resistance ?) or anything on the HT.
Not to mention the original wiring don't have safety earth  ;D

If it's confirmed, I can't let this configuration wired and operational for future user...

IIRC the actual safety rules state that over 50V its considered "HT" and absolutely no part can be accessible by human, whatever power or amperage involved.

Best
Zam

 

[radardoug]

The current drawn will depend on the skin resistance of the person connected, but there is no way you can allow this voltage on a piece that people can touch. It could be lethal.

 

[zamproject]

The current drawn will depend on the skin resistance of the person connected, but there is no way you can allow this voltage on a piece that people can touch. It could be lethal.

Don't make me wrong ! I won't rewire this system as is

Just try to understand how they allow this, even 70 years ago with less restrictive regulation.
To me the amperage is limited by the 500k ?

Best
Zam

 

[zamproject]

Hello

The easy part is done, power supply and amp section (dedicated chassis) is working
All capacitor and carbon resistor have been changed, as the five tubes sockets
All measured voltages seem in realistic range, I'll do some better measure/math to check more precisely bias etc...

Best
Zam

 

[zamproject]

Hi all

Some little update
The Ondioline pass sound since 2 days
I was more than happy as it was the first power on for the whole assembly (the 3 block, amp/osc/keys)
I'm suspicious about few things, but I'm now in the full check process.
As for some functions it's difficult to know if it work properly I still have to figure out what precisely "should" do some part of the design

first power on

View attachment Ondio1.mov

so far so good
Best
Zam

 

[JohnRoberts]

You definitely should not be able to touch a wire with 140 volts on it! That will hurt! Are you sure its not some sort of capacitor arrangement? So the wire will be insulated, but press up against the other wire to change capacitance?

Not good practice but current limited to 300 uA will reduce 140VAC to a tingle and 140VDC to nothing much at all.

GFCI protection thresholds are on the order of 6 mA or 20x that current.

Of course that current limiting is a function of components that can fail. Be careful.

JR

 

[clintrubber]

Hi all
...

first power on

so far so good
Best
Zam

Nice!

Is that song 'Samba Struck' ? (as later covered by AC/DC as 'Thunderstruck' )

Samba Struck / Samba Struck - live by Mambo Kurt

 

[zamproject]

Of course that current limiting is a function of components that can fail. Be careful.

I will ! this part will not be rebuilt as original, I have two option, a separated and isolated 5V supply that swich a relay, we can touch 5V without insulation (I check safety regulations), or recycle the touch detection circuit I design for my fader automation which can drive a 5V relay with simple modification

I already get some various relays (reed and standard) hoping one will be fast enough for this "percussions" circuit
IIRC the reed I have is about 1ms

Since I realise that the voltage is only about 40V when the percussion string is being activated (via switches) to drive the remote grid of output tube, there is a 75k inserted in parallel to ground meaining the divider is then 500k+65K
but the coper string stay at HT/2 when not inserted (500k+500k), and you still can touch it.

Is that song 'Samba Struck' ?

hahahaha !!

Best
Zam

 

[thomasdf]

Magnifique !

 

[zamproject]

Hi all

Some update
After troubleshooting most issue, everything work as I suppose it should.
The two sub oscillators are just amazing...never heard such bass syth !!!
I still have the "percussion" chord to wire with safe system.
Also hum optimisation if I can...

I have a question regarding the power amp push pull and phase inverter adjust (see attached)

In my unit the inverter is a 12au7 in place of the 6SN7, otherwise it's exactly the same layout.
My method was to scope both 6F6 grid in A+B mod, and adjust the volume pot (from first triode plate to second triode grid)
to get less possible signal (as I sum out of phase signal)
Is that correct method ?

Also if someone have a recommendation to add a direct out ? after the 6J5 ?

Cheers
Zam

 

[clintrubber]

Also if someone have a recommendation to add a direct out ? after the 6J5 ?

Or perhaps simply after the output-TX? (adding resistive divider to secondary, across speaker, and add jack)
Like for instance here (top right of schematic)

https://www.ampwares.com/schematics/twin_reverb_sf_135.pdf
Or make it a bit more fancy with a switching jack - when a jack is inserted (for line-level direct out from the divided-down secondary signal), the speaker is replaced by a load-powerresistor)

 

[abbey road d enfer]

Hi all

I have a question regarding the power amp push pull and phase inverter adjust (see attached)
View attachment 83924
In my unit the inverter is a 12au7 in place of the 6SN7, otherwise it's exactly the same layout.
My method was to scope both 6F6 grid in A+B mod, and adjust the volume pot (from first triode plate to second triode grid)
to get less possible signal (as I sum out of phase signal)
Is that correct method ?

Not really, because it doesn't take into account possible (probable) unbalance between the output tubes. Th ecorrect method is finding the point where the distortion at the output is minimum.
One thing nags me. There should be a resistor before the 500k pot, to compensate for the extra gain of the second triode. Actually, the attenuation is that of the pot, and its position must be very close to the start of travel, giving poor resolution. I would insert a 1Meg resistor there.

 

[zamproject]

Or perhaps simply after the output-TX?

Don't know yet If I want the output stage in the DO sound...
Maybe I should try various options and see what's most useful

I would insert a 1Meg resistor there

This just to have the pot 3x more precise ? (1/3 from 1M5)
Then should I also replace the other side grid 500k for 1M5 ?

I will try a REW with a mic at the HP... and check THD, and see what's possible with this setup.
(I suppose I'll need to feed a pure sin to the power amp block... the ondioline oscillators are full of harmonics and I suppose it will be difficult to track lower distortion)

Best
Zam

 

[abbey road d enfer]

This just to have the pot 3x more precise ? (1/3 from 1M5)

Correct.

Then should I also replace the other side grid 500k for 1M5 ?

No. The grid this side sees the wiper of the pot, so probably much less than 500k.

I will try a REW with a mic at the HP... and check THD, and see what's possible with this setup.
(I suppose I'll need to feed a pure sin to the power amp block... the ondioline oscillators are full of harmonics and I suppose it will be difficult to track lower distortion)

Correct. I think it should not be too difficult since there is an input connector.
Regarding DI, I think you should completely by-pass the amp, use a standard active DI.

 

[zamproject]

No. The grid this side sees the wiper of the pot, so probably much less than 500k.

I mean the 6F6s grids not the second triode of the 6SN7
One see a 500k (resistor) other the 500k (pot)
If I add 1M in series at pot side should I do the same for the other side ?
Or maybe I just miss something ?

Bypassing the amp was probably my first thinking for the DO, the "trouble" being that part of the Ondioline also have the PSU for the whole syth, I can't disconnect it, at best I'll have to clamp input to ground (if I don't want sound from the HP), but I suppose I need to keep all tube under power to ensure proper voltage all around (12 tubes...).

By the way, I think the OA2 that feed the 12AU7 oscillator tube is not working properly, I have more than 160V
But that's another story, I'll think I have to spare all tubes to be safe, as maybe switching the power amp for other tubes as 6F6 don't seem that common

Best
Zam

 

[abbey road d enfer]

I mean the 6F6s grids not the second triode of the 6SN7
One see a 500k (resistor) other the 500k (pot)

No, one sees a fraction of the pot. Note that there is a mistake in the schemo. Either the grid leak resistor for the bottom 6F6 is missing or the 100nF is superfluous. The grid needs galvanic connection to ground.

If I add 1M in series at pot side should I do the same for the other side ?

No.

Bypassing the amp was probably my first thinking for the DO, the "trouble" being that part of the Ondioline also have the PSU for the whole syth, I can't disconnect it, at best I'll have to clamp input to ground (if I don't want sound from the HP), but I suppose I need to keep all tube under power to ensure proper voltage all around (12 tubes...).

Bypassing it, yes, not shutting it down. Indeed, correct operation depends on the whole thing being on.

By the way, I think the OA2 that feed the 12AU7 oscillator tube is not working properly, I have more than 160V

Is it stable? That's the main point here. It's not that far off, for a 50 yo tube.

 

[zamproject]

tks Abbey !

No, one sees a fraction of the pot. Note that there is a mistake in the schemo. Either the grid leak resistor for the bottom 6F6 is missing or the 100nF is superfluous. The grid needs galvanic connection to ground.

Good spot, The 100nF at bottom 6F6 is not installed in my unit, that's why I'm thinking the pot is the grid leak resistor ? and the wiper (fraction of the pot) only go to inverter grid (which plate go to top 6F6 with 500k leak)
I'm still not confident with tube, so I probably miss something in your statement.

Is it stable? That's the main point here. It's not that far off, for a 50 yo tube.

70 years old !!!
Can't say for sure I just check few time after 20-30 min and seem consistent about 162V IIRC
It's definitely at the todo list to monitor this in a better way

I run a rapid REW setup for THD measurement, amp section directly feed by a sin and my B&K mic in front of the speaker.
I need to study more the best target figure for this but for the moment I was able to set h2 and h3 both at aprox -50dB relative to fundamental, giving about 0.50%THD in REW. The hum is in top of that by far (need to fix it...) and the noise is huge, but I don't have anechoic chamber

Best
Zam

 

[abbey road d enfer]

I need to study more the best target figure for this but for the moment I was able to set h2 and h3 both at aprox -50dB relative to fundamental, giving about 0.50%THD in REW.

That is not bad. What is the position of the 500k pot?

The hum is in top of that by far (need to fix it...) and the noise is huge, but I don't have anechoic chamber

Why an anechoic chamber for measuring electrical noise?

 

[zamproject]

That is not bad. What is the position of the 500k pot?

Top to bottom the pot measure 700k (for a 500k label...)
I read 79K at the wiper so just over 10%
I should probably better use a 100k pot in series with a 500k (or 200k with 1M) to optimise fine tuning accuracy ?

Anechoic is more a joke, but by listening in my "quiet" room I'm sure part of the background noise was also due to the transformer noise mechanical coupling (which was close), as I say I use a B&K (omni) which pick anything
I definitely have to improve hum, when I check/set THD at low frequency (close to 50Hz or upper harmonics) intermodulation beat a lot. 50Hz is 30 db below signal when I set THD... louder than h2 and h3 and with a noise about -60dB it's tricky to adjust harmonic peak in all that noise.

Now (or tomorrow..) I will check what I have at scope with my previous method.
I'll see how the drift is at pushpull grid "null" test if THD is set at output like you recommand me to do.

Best
Zam

 

[trobbins]

The schematic doesn't show the output transformer secondary is grounded. I'd suggest trying to probe the speaker output and feed that to your soundcard to then use with REW - that approach gives a much better tool for testing all manner of performances such as hum fault-finding and reduction, and frequency response, and signal clipping, and harmonic assessment of tones and vibrato/trem effects. You may need a resistor divider on the speaker output to reduce that max level to where the soundcard doesn't clip, and a laptop and modern USB soundcard make an excellent bench tool.

The heaters likely need a humdinger pot to gnd.