IF Amplifier input impedance value ???

"sparks"

** I reckon there is very little IF signal appearing across R57 (150 ohms) as the primary Z of T13 is much higher. So, the input Z is simply R54 plus R57 ( ie 300ohms) in parallel with the input Z at the base of VT13.

( VT13 = RCA 40235, Hfe =170, Ft = 1GHz )

But VT13 has an unbypassed emitter resistor of R58 (27 ohms), so its base Z is many times higher.

So, the load as seen by the Xtal filters is fairly close to 300 ohms, say

250 ohms.

Seems a not unlikely value for Xtal filters.

..... Phil

"sparks" wrote in message news: snipped-for-privacy@v13g2000pro.googlegroups.com...

View in a fixed pitch font:

150 150 Vin Iin ___ V2 I2 ___ -------->---|___|----o--->--|___|----. | | | | / \\ | ( | ) Ic=Vin/27 | \\V/ | | | | | === === GND GND

Iin = Ic + I2 = Vin/27 + I2 = Vin/27 + V2/150

I2 = V2/150 = (Vin - 150*In)/150

2*Iin = Vin/27 + Vin/150

Vin/In ~ 50 Ohms

The LTSpice simulation below seems to agree with this.

Do those filters contain internal matching transformers?

Version 4 SHEET 1 880 680 WIRE -80 -32 -224 -32 WIRE 112 -32 -16 -32 WIRE 288 -32 112 -32 WIRE 448 -32 288 -32 WIRE 688 -32 528 -32 WIRE 288 0 288 -32 WIRE -224 16 -224 -32 WIRE 112 16 112 -32 WIRE 688 16 688 -32 WIRE 288 112 288 80 WIRE -224 160 -224 96 WIRE -96 160 -224 160 WIRE 48 160 -96 160 WIRE 112 160 112 96 WIRE 112 160 48 160 WIRE 224 160 112 160 WIRE 688 160 688 96 WIRE 288 240 288 208 WIRE 48 288 48 160 WIRE 656 336 496 336 WIRE -224 352 -224 160 WIRE 288 352 288 320 WIRE 288 352 160 352 WIRE 496 368 496 336 WIRE 656 368 656 336 WIRE 288 400 288 352 WIRE 160 416 160 352 WIRE 496 480 496 448 WIRE 576 480 496 480 WIRE 656 480 656 448 WIRE 656 480 576 480 WIRE -224 528 -224 432 WIRE 48 528 48 368 WIRE 160 528 160 480 WIRE 288 528 288 480 WIRE 576 528 576 480 FLAG -224 528 0 FLAG 48 528 0 FLAG 288 528 0 FLAG 688 160 0 FLAG 160 528 0 FLAG 576 528 0 FLAG -96 160 Vin SYMBOL res 640 352 R0 SYMATTR InstName R1 SYMATTR Value 4k7 SYMBOL res 544 -48 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R2 SYMATTR Value 150 SYMBOL res 272 224 R0 SYMATTR InstName R3 SYMATTR Value 27 SYMBOL res 272 384 R0 SYMATTR InstName R4 SYMATTR Value 220 SYMBOL res 96 0 R0 SYMATTR InstName R5 SYMATTR Value 3k3 SYMBOL res -240 0 R0 SYMATTR InstName R6 SYMATTR Value 150 SYMBOL res 32 272 R0 SYMATTR InstName R7 SYMATTR Value 3k3 SYMBOL cap 144 416 R0 SYMATTR InstName C1 SYMATTR Value 100n SYMBOL cap -16 -48 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 32 32 VTop 0 SYMATTR InstName C2 SYMATTR Value 100n SYMBOL current -224 352 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName I1 SYMATTR Value SINE(0 1u 1.4e6) SYMBOL voltage 688 0 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 12 SYMBOL ind2 272 -16 R0 SYMATTR InstName L1 SYMATTR Value 5m SYMATTR Type ind SYMBOL ind2 480 352 R0 SYMATTR InstName L2 SYMATTR Value 5m SYMATTR Type ind SYMBOL npn 224 112 R0 SYMATTR InstName Q1 TEXT 406 160 Left 0 !.tran 5u TEXT 408 88 Left 0 !K L1 L2 1

The above ignores re which is about 2 ohms, since Ie ~ 14mA.

This gives Ic = Vin/29 which gives Vin/Iin even closer to 50 ohms.

A surprising result for matching crystal filters.

I'm not sure how you can say anything about the primary impedance of T13 without data that's not in the schematic -- it _could_ have a high turns ratio, or that line going off screen could be doing something. Certainly the designer thought there would be some signal at the cold end of T13's primary or C34 wouldn't be there.

But 250 - 300 ohms sounds good for an Xtal filter.

I'd be tempted to rig up a test set and sweep the filter for its response, regardless of what I learned from analyzing the circuit. This isn't hard to do by hand (at 1.4MHz you can pretty much get away with protoboard for the passband portion), and it gives you the opportunity to make your own impedance/ripple/skirt steepness tradeoffs.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

Thanks for the help on this guys. In the interim, Ive had some advice from someone who used to work at Redifon (the radio's manufacturer) and he says the filters they used were sometimes (but not always, depending on radio model) 50 ohms.These filters used small internal matching transformers inside the case. There were other options however for filter value such as

280 and 680 ohms. So it looks like the filters are 50 ohms by your calculations.... Thats great confirmation. regards, JEFF

ply

ly

s .

d text -

p n

imply

f 3 s

inly

s

his

to

ys.

ed text -

Hi Guys, Ive placed another circuit of the IF Amp on box.net, which contains the missing diode atten network. Its here...

I have another 2 related questions on the design I wonder if you could help me with please ...

What do I need to change in the circuit so to increase the input impedance to 910 ohms (so I can use another more modern brand of filter) ? Would this alteration to 910 ohms still be consistent if the IF freq was moved up to 10.7Mhz, as this is the preferred filter freq Id like to use.

regards, JEFF

Amp

he

EI

on

a

ck

or

0

simply

of

s

s,

T13

rns

tainly

3's

=A0This

y to

says.

ted text -

"sparks" top-posted:

Remove R54. Reduce R55 and R56 to 1K8.

Should be. T13 might be less efficient at 10.7 MHz.

Are there any tuned circuits in the later IF stages? What about the BFO? Changing IF frequency sounds a bit drastic.

CORRECTION: Just tried this in simulation and it doesn't work. This is better:

Leave R55 and R56 alone. Remove R54. Short the cold end of C34 to ground and/or short-out R57

Thanks Andrew. Im wanting to use a wider filter and hence the change to 10.7Mhz IF where I can obtain suitable (free) 8 pole filters of various bandwidths. These filters are 910 ohms, hence the question re changes required to accommodate the 910 termination impedance. Its probably likely that the transistors used in the IF wont be too good at 10.7Mhz, but I havent checked their spec yet. Yes, there is a couple of tuned circuits in the last part of the IF. Noise shaping the manual says they are for. The first is balanced output to accommodate a balanced crystal filter config for CW reception, but I plan to gut it, and use only one tapped tuned circuit to match into the detector. BFO is present but not required. How easy / expensive is LTSpice? Ive heard of it, but never seen it in action. JEFF

LTSpice is easy to use and free!

40235, 2N3904 and 2N3906 should be fine at 10.7 MHz