In Reply to: Some preliminary results posted by mfc on July 19, 2003 at 10:19:55:
Here are some interesting results for the MLJ3281 Sizklai follower
with compensation. Square wave at 8v peak. 3900pF capacitor inserted
between collector and base of first transistor.Square Wave = 1050 Hz. Using 1,3,5,...25,27 harmonic values only.
27th = 28350Hz. Amptitude = 8v peak.
Gibbs phenomena (although present)was very small.
Sine Wave = 5KHz. Amptitude = 1v peak.B+ = +-20v, ILoad = 1.9477 amps, RLoad = 10 ohms
Vbase = 14.6mv.Simulation settings: Step = 100nS, Window = .1second
Driven with a 1V peak sine wave only and measured output(1st column)
Compensated and Driven sine/square wave measure output (2nd column)
Uncompensated and Driven sine/square wave measure output(3rd column)
The noise floor on input was about .3nV which represents a
very low noise level.Here are the results:
Harmonic VRef Vout VoutFndmntl(dB)0.00 0.00 0.00
2nd -87.50 -57.61 -57.10
3rd -118.33 -61.53 -61.44
4th -133.76 -70.11 -71.42
5th -136.77 -81.06 -81.13
6th -136.42 -78.44 -79.20
7th -135.45 -85.52 -91.51
8th -136.10 -86.19 -87.47
9th -136.62 -88.14 -90.89
10th -135.55 -91.52 -100.22
11th -136.86 -96.35 -96.35
12th -136.72 -89.10 -91.89
13th -135.61 -86.04 -99.84The compensating capacitor has some effect on individual
harmonics. However, compensation actually makes harmonic levels
worse. Overall, they are slightly worse with the capacitor than
without it.The levels are no where near the very nice levels of the uncompensated
4v square wave. This indicates that the square wave at this level
(8v) is starting to show some slewing, and the compensating cap just
makes it a little worse.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Compensating Sziklai doesn't help (that much) - mfc 07/22/0316:07:24 07/22/03 (16)
- Re: Compensating Sziklai doesn't help (that much) - Scott Frankland 18:13:57 07/22/03 (15)
- Re: Compensating Sziklai doesn't help (that much) - mfc 00:15:01 07/23/03 (14)
- Re: Compensating Sziklai doesn't help (that much) - Scott Frankland 00:51:43 07/23/03 (13)
- Cap loaded - mfc 12:28:03 07/23/03 (12)
- Re: Cap loaded - Scott Frankland 12:55:44 07/23/03 (11)
- Re: Cap loaded - Scott Frankland 13:47:34 07/23/03 (10)
- Re: Cap loaded - mfc 14:18:30 07/23/03 (9)
- Re: Cap loaded - mfc 15:19:39 07/23/03 (8)
- Re: Cap loaded - Scott Frankland 17:41:41 07/23/03 (7)
- Re: Cap loaded - mfc 00:36:32 07/24/03 (6)
- Re: Cap loaded - Scott Frankland 11:43:40 07/24/03 (5)
- Re: Cap loaded - mfc 17:20:42 07/25/03 (4)
- Re: Cap loaded - Scott Frankland 22:22:30 07/25/03 (3)
- Re: Cap loaded - mfc 12:13:58 07/26/03 (2)
- Re: Cap loaded - Scott Frankland 13:02:21 07/26/03 (1)
- Re: Cap loaded - mfc 14:58:45 07/26/03 (0)