22-11-2012, 11:16 PM
Designed and built way back in 1982 (that's why it looks a bit shabby) - but still working. 
[attachment=6358]
Sine, square and TTL outputs: sine & square outputs calibrated for output up to 1-v. r.m.s. into 600 Ω. Freq. range 10 Hz to 1 MHz; O/P substantially 'flat' over that range, but with about a 3 → 4 dB drop on the highest freq. range.
Mains powered; uses three LM 318 op-amps.; LM 316; SN7406N; LM340T-5; LM340T-12; LM320T-12. Basic osc. is a Wein bridge with 2N5457 F.E.T. for amplitude stabilisation.
[attachment=6359]
[attachment=6360]
No precision components are involved, but some are 'select-on-test' and there are a few calibration pre-set resistors. The one (now) difficult-to-get component is the 5 kΩ log. / reverse log. pot. which is used for the variable freq. control. That item is the weak point in this design, since ordinary, commercial pots. of that law are far from being true to that law. But at the time, that item was what was available - and it enabled the design to meet my required spec.
The waveform is not without some distortion - but the sinewave output is quite adequate for signal injection for basic testing of audio amps. and low-freq. R.F. stuff.
[attachment=6361]
And the TTL O/P is obviously useful when testing logic circuits. The freq, scale is approximate - but then I always use this generator with a 'scope and a freq. counter, and it is not often that I require to know the exact freq.
Stability is adequate: needs about 10 minutes to really 'settle down'. I am yet to determine the long-term freq. stability, but up to now it hasn't caused me a problem.
Al. / Nov. 22, 2012 //
[attachment=6358]
Sine, square and TTL outputs: sine & square outputs calibrated for output up to 1-v. r.m.s. into 600 Ω. Freq. range 10 Hz to 1 MHz; O/P substantially 'flat' over that range, but with about a 3 → 4 dB drop on the highest freq. range.
Mains powered; uses three LM 318 op-amps.; LM 316; SN7406N; LM340T-5; LM340T-12; LM320T-12. Basic osc. is a Wein bridge with 2N5457 F.E.T. for amplitude stabilisation.
[attachment=6359]
[attachment=6360]
No precision components are involved, but some are 'select-on-test' and there are a few calibration pre-set resistors. The one (now) difficult-to-get component is the 5 kΩ log. / reverse log. pot. which is used for the variable freq. control. That item is the weak point in this design, since ordinary, commercial pots. of that law are far from being true to that law. But at the time, that item was what was available - and it enabled the design to meet my required spec.
The waveform is not without some distortion - but the sinewave output is quite adequate for signal injection for basic testing of audio amps. and low-freq. R.F. stuff.
[attachment=6361]
And the TTL O/P is obviously useful when testing logic circuits. The freq, scale is approximate - but then I always use this generator with a 'scope and a freq. counter, and it is not often that I require to know the exact freq.
Stability is adequate: needs about 10 minutes to really 'settle down'. I am yet to determine the long-term freq. stability, but up to now it hasn't caused me a problem.
Al. / Nov. 22, 2012 //
