Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

OK, so my signal generator has gone sick so butchering it with some upgrades and mods.

Status
Not open for further replies.

Boxnut

Member
The triangle generator looks more shark fin so gutting and replacing that section. RF oscillators need help but have to wait on my spectrum analyzer for that.(another project)
A mod in mind is adding like a dip meter to find resonances. Given I need to amplify the emitter current anyway, a thought of using an op-amp. a long time since working with those. I don't want to violate the broadband nature of the output amp and moot for someone with my background. What is new to me is the amp is minus V powered and driving me crazy solving how to go about this. The goal is DC coupled to a scope and ruled out isolating all this. Referring to the attached file, I want to measure and amplify the current across R61. I have some 741s and some older C4558s that should work for this. Now--Does anybody have insight how I might add this in some detail.
Any help here would be hot, thanks in advance.
Keep in mind the emitter is above ground(-18V)
 

Attachments

  • U9 PA 1060.pdf
    76.3 KB · Views: 180
Triangle wave distortions at the peak (near the peak) can be caused by your probes. Too much current load from a 1x scope probe or a grounding clip that is too long can cause strange effects.
 
there's about 3V across R61, or 44mA through the resistor (subtract the B-E diode drop from the measurement across R59). you can AC couple the scope and see what the signal looks like there (using the chassis ground for the scope probe ground)... the -18V rail has DC on it, but as far as AC goes, it's still considered an AC ground. alternatively you could use both channels of the scope in subtract (differential) mode, and connect one probe to the -18V rail, and the other to whatever you are measuring. you need to keep the vertical gain selectors on the same setting. the way to set it up is scope input "A" is put into invert mode and the vertical mode switch to "Add". then connect both probes to the -18V rail, and with both input channels set at 5V/cm. your trace should be at or near the 0 line. now whenever you move the "B" channel probe to a test point the scope is displaying the difference between the two probes.
 
Last edited:
@ unclejed613. " there's about 3V across R61, or 44mA through the resistor " Yes, that is about right with no signal, the bias current for the amp stage.
With signal, this goes to about 5 V. Class AB1 or about.
AC across R61 tells me little and frequencies way above the scope limit (100 Mhz)
The impedance for this amp is 50 Ohm. Now say I connect a random length antenna to the output sweeping it, there is some frequency point where it is resonant/ coupling max. The output amp draws minimum current at this point, like a dip meter. 'THAT' is what I want to look at and the scope is tied up looking at the sweep and detected output. The spare channel, I could find this dip and guestimate the resonant load. Finding the exact frequency with my spectrum analyzer. My thought was isolating R61 with a DC op- amp, simple with a B+ rail. My headache here is the B- rail. My scope will see the 18V fine, an older Tektronics 465. I will solve this in time, just that op-amp applications are not my strong point and many cups of coffee later, still uncertain.
Can't be that hard. Thanks for the input.
 
Triangle wave distortions at the peak (near the peak) can be caused by your probes. Too much current load from a 1x scope probe or a grounding clip that is too long can cause strange effects.
I am gutting that portion of the circuits, that awful waveform is reflected in the generator output. This is a 35 year old Wavetek, prone to fail just from age.
I have a nice generator module, good clean square for blanking and a good triangle. even has the 5 us holdoff adjust I need.
 
you could tap off of both sides of R61 with a pair of 10k resistors. the "bottom" resistor, you would couple to ground with a capacitor (1nF or more should work) and the "top" resistor goes to the oscope input. this would give you a very light load on the circuit, yet give you a good relative measurement for monitoring the current dips.
 
on second thought, the "bottom" resistor is probably not necessary since that side of R61 is already bypassed to ground...
 
on second thought, the "bottom" resistor is probably not necessary since that side of R61 is already bypassed to ground...
Yes, you also have to consider this is a very broadband amplifier and changes here could put a ding in that. One thing I am hoping to avoid.
Bandpass for this is 50 Khz to 600 Mhz.
 
you could tap off of both sides of R61 with a pair of 10k resistors. the "bottom" resistor, you would couple to ground with a capacitor (1nF or more should work) and the "top" resistor goes to the oscope input. this would give you a very light load on the circuit, yet give you a good relative measurement for monitoring the current dips.
I had though of a couple chokes to isolate any RF component across R61, I don't think the scope probe would be very reactive, but back to my issue, the probe needs to be grounded and this is all at or near 18V minus. The dips I am guessing will be a volt so easy to see with the scope. DC isolation is what I am hoping for and even a unity gain amp should do the trick. AC coupling should see the dips while sweeping but that would change manually fine tuning the frequency.
 
you don't want to add any long wires to the emitter connection.... best to add the resistor right there at the transitor, and then any long wires from the resistor are isolated from interacting with the emitter circuit
 
you don't want to add any long wires to the emitter connection.... best to add the resistor right there at the transitor, and then any long wires from the resistor are isolated from interacting with the emitter circuit
Of course, the notion behind the chokes but I will try the resistors. still working up a breadboard for something to isolate the DC and still eyeing a discreet device or op-amp. This is housed in a can so have to keep everything close to the transistor.
Thanks.
 
you could tap off of both sides of R61 with a pair of 10k resistors. the "bottom" resistor, you would couple to ground with a capacitor (1nF or more should work) and the "top" resistor goes to the oscope input. this would give you a very light load on the circuit, yet give you a good relative measurement for monitoring the current dips.
Hey, I tried the resistors and that looks like it will do the job. Still could use some amplification though as not much swing across R61.
Solved some oscillator problems but still have like a 2Khz jitter or waver as if being FMed. 2 detector diodes to replace, one is the leveler sense, the other as detector for the scope vertical. Not cheap at $25 each. I will let you know how this pans out when completed, Thanks much for your suggestions.
 
Still could use some amplification though as not much swing across R61.
that's because the transistor is an emitter follower... what's being amplified there is current, not voltage.
Not cheap at $25 each.
what are they, the cartridge diodes they used to use for RADAR? when i was a teenager, i had a large collection of them.... i wish i still had them they seem to be worth their weight in gold.
 
that's because the transistor is an emitter follower... what's being amplified there is current, not voltage.

what are they, the cartridge diodes they used to use for RADAR? when i was a teenager, i had a large collection of them.... i wish i still had them they seem to be worth their weight in gold.
First part,,nope this is the output power amplifier, last of 3 stage.
Second part,,not quite just low IMD and 10 Ghz upper range. hard to find and just not cheap. Solving the flatness leveler means at least one, investing in an active power sensor from Agilent, so I can work with that.
The diodes in there now roll off above 600 Mhz.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top