redrawing TridentCB9146

[abbey road d enfer]

In tech doc found a D&R mono compressor.... Similar design to Trident ... Except it has a mirror Jfet at vca...
Is it used to better bias the first one and distortion cancelation?

No, it's just that it allows halving the resistors at the input, which results in slightly lower noise.

 

[richiyobs]

Fine... Thanks for explaining

 

[richiyobs]

U2b is the right way to leave the half part of an unused opamp?

 

[abbey road d enfer]

Yes, that's the commonly recommended way, although I once had a problem with an opamp that went short between one rail and the non-inverting input of the unused amp, which resulted in shutting the whole unit. That's why I uually add a 10k resistor between the non-inverting input and ground.

 

[richiyobs]

....10k resistor between the non-inverting input and ground.....
thank you so much!
last question, can i use this to build a psu?
https://www.reichelt.com/fr/en/tran...i-54-18-8-215-p27380.html?&trstct=pos_3&nbc=1or can you please advice me any alternative?

rectifier 1n4002 and regulators will be 7818 & 7918

thanks in advance for your great support

best

 

[abbey road d enfer]

....10k resistor between the non-inverting input and ground.....
thank you so much!
last question, can i use this to build a psu?
https://www.reichelt.com/fr/en/tran...i-54-18-8-215-p27380.html?&trstct=pos_3&nbc=1or can you please advice me any alternative?

rectifier 1n4002 and regulators will be 7818 & 7918

What current do you need.
The xfmr delivers 15Vac at 533mAac, which would result in about 19.5V dc after rectification. that would not be enough for proper operation of the regulators (Vout+3V). However, the open load voltage is 18.7, which results in nearly 25V after rectification. That would leave about 7V for ripple.
In short, this xfmr will work for moderate current draw, probably about 150mAdc max. If you intend on drawing more, you should either used nthe 18V version of transformer, or use 7815/7915.

 

[richiyobs]

Still have to learn how to evaluate pwr consumption....
I choose dual 15 cause I builded one psu with 7x18 regulators and it worked...but could not evaluate Ripple
I often see 10vA xfmr in diy projects

The original Trident schemo has 100mA fuse at xfmr input
So how to measure circuit pwr Consumption? From Tfx secondary to psu board or after regulators to entire circuit?
Sorry for my stupid question... Just to be sure

Best

 

[abbey road d enfer]

Still have to learn how to evaluate pwr consumption....

The best way is to evaluate it from the schemo, which must be complete with every option and LED and whatnot.

I choose dual 15 cause I builded one psu with 7x18 regulators and it worked...but could not evaluate Ripple

Probably had low power draw.

I often see 10vA xfmr in diy projects

Because you can see many cars with a 1.6l engine does not mean that all cars can run with it.

The original Trident schemo has 100mA fuse at xfmr input

It means that the actual power is less than 24VA, most likely about 10-15. Does not help much.

So how to measure circuit pwr Consumption? From Tfx secondary to psu board or after regulators to entire circuit?

You can do both but measuring from the secondary is not very effective because the current there is very distorted and a typical DMM will not give an accurate value.
Measuring DC current is better, but you must multiply it by 1.414 for estimating the AC power.
But since it supposes a fully built unit, it is better to measure the unregulated rectified voltage with an oscilloscope, and make sure that the minimum is still at least 3V above the desired regulated voltage.
Another way is simulating the PSU with LTspice, but it implies having an accurate model for the xfmr, including DCR and leakage inductance, which is almost never the case.
You have to know that using a 2x15Vac xfmr for producing +/-18Vdc is somewhat precarious.

 

[richiyobs]

Thanks again for all the infos

 

[richiyobs]

attached a psu simulation i m running
not sure about tfx... just changed inductance value to get a working voltage

 

[abbey road d enfer]

attached a psu simulation i m running
not sure about tfx... just changed inductance value to get a working voltage

Many things are wrong in your sim.
First voltages in spice are peak, so the mains voltage source should be 325V (230 x sqrt2).
You have not correctly evaluated the inductance ratios. For 230:18.7V you need a ratio of 500:3.06, not 100:1
Then you have modelled the nominal inductances of the xfmr (waaaaay too low but it doesn't matter much - the xfmr would explode in the real world), but you haven't modelled the leakage inductance (or alternatively edited the coupling coefficient K for 0.99xx).
Such a xfmr usually has a primary inductance of several hundred Henries, and the leakage inductance is about 1/1000th of the nominal.
You have not included the DC resistance of the windings.
These are the most important factors in evaluating the rectified voltage.
DCR are easy to measure, but leakage inductance is not so easy, because a DMM or LCR meter usually don't have the juice to test the xfmr in conditions close enough to nominal operation.
You haven't attached the models for 7818 and 7915.

 

[richiyobs]

78xx models ... sorry forgot them...
i was very lucky to make sim works, also beacuse i really don t know how to dermine and include the xfmr tech spech ... no tech data on pdf referred to linked one

 

[richiyobs]

Finally builded and tried to calibrate this device....
Everything went fine except I spent lots of hours setting the meter...
I got this: with no signal moving Ratio to max or RELEASE to Slow, the needle moves to - 20
Noticed that biasing more the jfet and finding a balance between meter trimmer this excursion is reduced a lot....
Is this normal for this kind of meter driver?
The meter I m using is Sd39 500uA.... The driver part seems to have a wide range of use, can be easily adapted
Respect to original design I m using 1n34 instead of original germanium ones
Thanks for your attention

 

[richiyobs]

Got it! The meter sens trimmer was damaged...
And so a bad behavior...
Attached the proto picture
Best

 

[richiyobs]

running the device in a mix it was noisy and found that IC9 has a lot of gain....
components close to this opamp act like a pick up!
so trimmed it down to a usable range....now too low respect to calibration procedure, noise disappeared but C17, 20 and 21 still produce pick up fx
is it maybe caused by my layout arrangement, or is IC9 really high in gain?
thanx for attention
best

 

[richiyobs]

can someone please help me troubleshooting my device replica?
had time to etch a different layout, but same story, C17 C20 and C21 still produce Hum-buzz-noise when touched... even with 2n5458 not on pcb
i m not so expert to get it on my own...
Any advice please on what to check?
thanks in advance
best

 

[abbey road d enfer]

can someone please help me troubleshooting my device replica?
had time to etch a different layout, but same story, C17 C20 and C21 still produce Hum-buzz-noise when touched... even with 2n5458 not on pcb

All these caps are in the signal path. Touching anything that's in the signal path produces buzz.
Nothing wrong with it. Did nobody tell you "DO NOT tOUCH LIVE WIRES" ?

 

[richiyobs]

Thanks for clarifying... Yes I know to not touch live wires... But when I was close to them using trimmer for calibration... Noticed that effect
And was worried cause it s the first time I listened to that... Never noticed it into other devices! So started touching everything
Thanks again