Heathkit IG 18 Manual

Page 1

in which one of our redoubtaole
English cousins does a Professor
Higgins on Heath's metered sine-
square oscillator and turns an
Eliza Doolittle into a lady of
ravishing style and performance.

HOVLD You like to own a gen-
uinely high grade sine and
square wave audio generator! Just
look at these specifications:

mu: 1 Hz - )0sz sine or so.

RISE TUE: 12 nanoseconds

1.".D.: 000: typically, 0-002-
0-005!

ACCURACY: better than 20-53
APPLITUDE SlABlLlTY: 2 l 68.

Yes, I can hear you say, and
where do I rustle up the $400 or
more for what is obviously a labor-
atory standard instrument? well,
you lucky Audio Air-tour subscriber,
read on, for you can be the proud
owner of an instrument in this
class and for a bare fraction of
that price tag.

are of ly professional tasks is
evaluation of new products for the
most popular of English audio mga-
zines Ill-Fl News. Last year, l had
the good fortune to undertake a re-
view of the new Heathkit mdel IO-
18 audio generator. This transis-
tor version of their popular tubed
nodal lg-72 has identical perfor-
mance and ranges, but also includes
a square wave section.

Heath's specifications of the new
model are Iodest, with a claim for
less than o-lt distortion, 25 fre-
quency accuracy and a square wave

Copyright © 1971 by Edward T. Dell Jru'. All rights reserved.

By REGINALD
WILLIAMSON

rise tine of better than 50 nano-
seconds. In fact, a tolerably good
performance for something aired at
the service shop or the not-so-cri-
tical nateur. The kit had a couple
of emrs when I cane to check it
out, but these were corrected with
the willing cooperation of the
British cowany.(l'll be mentioning
these later on.)

In IV subsequent appraisal of the
product, therefore, i was pleased
to give it a warm welcome, with a
aild qualification about its total
harmnic distortion which it barely
let.

What about my warranty?

Later, however, while idly studying
the sche-atlc, I began to realize
Neath had given us a laboratory
standard instrument in embryo,
since the basic circuit design
philosophy was sound and needed
just a few refinements here and
there, to bring about quite a
startling performance inpmvcment.
To begin, you will need a Heath
lG-lB audio generator kit costing
$67.50. One word of warning, how-

;

FIGJ. Simplified schematic of the
operational amplifier Heath used in
the 16-19 design. The triangle- is
the amplifier: 1r; - input the in-
verting, the o the non-inverting.

ever. Quite rimtly, our friends in
Benton Harbor are perfectly en.
titled to invalidate their warranty
if the owner makes unauthorized
changes in their kit. Their excell-
wt construction manuals usually
make this quite clear. So first,
you should build the kit as the
manual directs and get it working
exactly as they specify. in this
way you can be sure the kit is free
of basic flaws before making chang-
es in it.

If subsequent to mdifying the
kit, you have trouble and want to
return it to Heath, then you .15!
be prepared to remove the modifica-
tions i shall be describing. 0n the
other hand, ln prepared to give
assistance to any Audio Amateur
mder who may write tu no in care
of the aagazine.

How these oscillators work

The :45, in common with mst au-
dio signal generators in use today,
is a resistance/capacitance (RC)
type, which means the frequency of
oscillation is deter-ined by sharp-
ly frequency selective resistance/
capacitance networks. Sonatiues.
the networks are in the bridge film,
as in the so-called uein bridge oa-
cillators; or, as in the case of
the 16-18, the frequency determin-
ing elements are in a "bridged '1"
configuration,

This network type sinulates a
broadly tuned resonant circuit, and
in the 16-18 is inserted in the de-
generative, or negative feedback
path of a high gain aapllfier. Neg-
ative feedback is at o maxi-um, ex-
cept at its "notch" frequency-do
if oscillation occurs, it can only
be at this frequency. So by singly
altering the values of R or C, the
frequency of oscillation is deter-
mined very accurately, to a toler-
ance defined by the accuracy of the
bridge consonants. What smell er-
rors that do occur, could be caused
by phase shifts within the aaplifi-
er; so this must be designed care-
fully and have a bandaidth well be-
yond the generator's frequency
coverage.

The operational are and loops

Heath's designers have cone up with
what is essentially, a very high
gain operational mlifier, with
low output impedance and two in-
puts, non-inverting and inverting.
The RC bridge is in the inverting
(or degenerative) path; the non-in-
verting path is the main regenera-
tive loop and is also the aaplitude
Controlling network.lSee Fig.1)

1 should explain at this point,
that because of the low "Q" of RC
tuned networks, it is essential to
include them in an active filter
configuration to insure low harmon-

issue 1/1971

7

Page 2

Copyright © 1971 by Edward T. Dell Jnr. All rights reserved.

8

it generation in an oscillator of
this type. It is highly desirable
that the forward anplifylnl path
should be as hlfli gain as possible.
but at all tics, irrespective of
the insertion losses of the fre-
quency determining networks. the
closed loop gain should be just
above unity.

The positive loop has in it a
positive temperature coefficient
thenlstor so as the aqlitude of
oscillations rises, it heats, re-
sistmce increases and the we
feedback is reduced, thereby close-
ly controlling the closed loop gain
at all frequency control ele-mt
settings.

Checkout before aodlfication

As it stands, the [6-18 checked out
well on this count. Over- the entire
range 1 Hz up to mom; it not
their specificatlon of aid! varia-
tion. Diatortlm wasn't quite as
good as specified, typically around
O-lZ to 0-148, rising appreciably
at the ends of the spectrum. liar-
Iaenlc products were a disconcerting
aixture of Iainly second, third,
and higher orders. lira also contri-
buted to the residue ehen the pro-
thlcts Here checked in an oscillo-
09'-

Accuracy was poor at so. points
md while even the 55 isn't too
render-(m for a decade instru-nt,
at same scale settings the 16-18
was 125 out! Finally. the urn
drove us and at frequencies below
min, when it started to read indi-

H62
lotus edmltinllevmltm

lu Iv

I I)

l "Ge
1 Me
l U )l
I Ha

Ill!
ll la
1 II It
I I! St
! ll H
II III

In

9-.

. an

vidual half cycles in and needle
oscillations, becoming virtually
useless in these ranges.

lbdlfi catimt: first step

i decided, first of all, to deal
with the set-lam bleaiahes--the
frequency inaccuracies. This turned
out to be I: error in settling
the kit--very rare for Heath, in 1
experience. Heath lqn-oved the IG-
lfl over its ancestor, the [672 by
previdlnl a continuously variable
third significant flare, a mat
useful feature when one is cheating
resonant cirwlts. The change vir-
tually removes the major- disadvan-
tage of I switched decade type in-
stmmnt.

Unfortunately, the design re-
quires the extra resistive elewt
to be switched out altogether
in the "0" position. The ordinary-
gmged potentioneter used aade lt
lmosslble to achieve this condi-
tim. So when the resistive (DQO-
nents in the bridge were relatively
hiflr in value, an error- of up to
lot was inevitable.

What Heath should have used was s
rheostat. After their attention had
been drawn to this, I was assured
the error has been corrected. But
if you have aade the kit already or
will be getting one, it is 3 else
precautim to check that this col-
parent»: 1 negohm ganged control,
It, in the Heath ache-atic~-does go
open circuit when in its zero or
fully anticlockwise posltlm.(See
Fig. 2.)

Copyrlgmolssa. Tn. lhath to. Used By per-lealon.

null!
latent-(v
tent-r

---..

lltaulntr
IOIIIDI

1 \
llllo

Haj. Heath dealgnera added a variable rhea-tat to the three witches of
the oscillator. ma dual 1N0 R8 goes open circuit in its minimum poaition.
Author "ulna-sen says the unit works better i! the cam resistors which
aeath supplies to go across the rheoetats teminala are replaced with
loom units. The control adds a variable one cycle to whatever- frequency

the network's three sud tchea produce.

The Audio Meteor

Sup two. the other a-ll layout
error Heath has also accepted, and
corrected, was glvlng rise to ex-
cessive crosstalk between the
square and sine wave outputs. All
that was necessary, was to ahift a
com (retold return for the sine
end square output attenuator sec-
tims. The are for the square wave
section is not separate and an ex-
tra wire is taken free the attenu-
ater dlnct to the mind line on
the wave generator etched circuit
board. All Heath's meals subse-
quent to Dec. 5, 1969 are aunded.
You'll find the date of issue at
the front cover, bottoa right.

F-N

IODIFICATION PARTS LIST
Step one:
I Dual "in linear rheostat

(Check your kit before orderlng)
2 loom, :2: iv. resistors

Step two:
One short Imgth of hookup wire
Step three:

l S-auF 9 30V bead tantalum car
cltor

Step four:

I 10ch 0 6V electrolytic cap.
Step five:

UK!) N resistor fist

10KB {ll resistor 25:

lm ill resistor est

MP 8 |6V Electrolytic cm.
loOuF E loll Electrolytic cep-
hohoe transistor (nth)

Step slx:

.............

2 lOOOuF Q 6W Electrolytic cap.
Step seven

l 5.6!? lit resistor 15%

l 250.5 6 6V Electrolytic cap.

I Ben-unsum diode (male, mar)
Step eight

l 2N5087 (Hot) transistor
1 22mm 5H reslstor *5!

Step nine

1 1:0361 (RCA) translstor (Qt)
l lI036! (RCA) transistor (05)

Step ten
l UK ill resistor :51

fi-J

Page 4

Copyright © 1971 by Edward T. Dell Jnr. All rights reserved.

l0

Step 7--Danlplng some naggles

Before leaving the meter circuit, I
decided to do something about the
poorly damped mvenent and added a
low voltage working, ZSOuF. capaci-
tor across the meter tendons. The
meter action linearity also left a
little to be desired, due to the
nonlinearity of the recti tiers, 03
and ()4. This was corrected by add-
ing a 5-H: resistor and a genuine:
diode across the meter terminals as
well. Aloost any small diode will
do, provided it is a gerlaniun type.

Take care in adding the diode and
capacitor to connect the proper
polarities according to Fig. 4.
This tightens up the meter accuracy,
incidentally, to within 12 as ag-
ainst Heath's specified :St. The
master now reads without that vio-
lent needle wiggle that irritated
n: so lich, dawn to 1 Mt where the
fluctuation in reading is less
than ildll.

Step eight. 1 lncreased the current
gain of Q: by converting it into a
cowound pair, Le. adding another
transistor. The additional transis-
tor, Q3a, shown in Fig. 5, should
be a high h, low noise device
with 5 Va" of not less than 45V.
and mst, of course, be a pup. me
has a variety to choose from and 1
used a Vatoroll ZN5087.

You'll find it quite easy to add
the extra trunsistor, 03a to Q3.
Just unsolder 03': base lead (with
a heat sink attached) and then lift
it up out of the hole in the board.
Bend this up in a "U" shape. Push
the base lead of the (13- transistor
through-this hole, and solder.

The collector lead of on should
be put through the same hole as
that of the original 03's collector,
and also soldered. Now join the
base lead of Q! to the emitter lead
of 03:, along with one end of a
220k resistor. The other end of
this resistor goes to a convenient
sve(Be) point on the etched circuit
board. I found it a good idea to
drill a sun hole (158 drill) near
the we end of R11 and solder the
2201 to the same section of the
copper luninate.

Step nine. I changed the output
transistors, Q4 and Q5 to larger
chip types, substituting the most
inexpensive available: an R0 40361
for 04, and RCAs 40362 for QS.(See
Fig. 3.)

Step ten. Change R5 in the differ-
ential input from its original val-
ue of l-7hflto 47km.

The new specifications

These are about all the nodifica-

lions any home constructor is able
to do. The oscillator should now be

The Audio Amateur

FIG. 5. Details of adding a trans-
Iltor to 03 to like it a compound
pair. See text.

HG. Luise time of the square wove
section of the 16-18. Distance be-
tween gtaticule divisions-O-lusec.
The author says the tlny ring at
the top is test lead inductance.

FIGJ. Additional Circuitry for isolating the meter from the ascillator's
amplifier and for damping low frequency oscillations 1:: meter reading.

set up in accordance with the in-
structions on manual pages 41 th-
rough 43. Please keep one minor
point in nind. The adjust-ant of
the feedback control, R7, is now
ping to be rather sharp, and hence
a little more difficult to set.
Also, when adjusting the bias con-
trol, us, have the feedback control
set to the point where clipping is
just noticeable a: the -ve and ove
half cycles of the waveform. You
will achieve the best results by
setting the bias so these clipping
points are as symtrical as possi-
ble.

Now, after doing all this, what
are the results? You nay rest as-
sured, performance of your 16-18
will now be up to laboratory stand-
ard. The distortion products of ly
unit are t ically 0'00, with
noise contributing an equal snount
so that the total is always less
than 0-0. Accuracy, as K said
earlier is 20-. as the instiulent
stands. If you can pt the use of a
digital counter, you could do as l
have and with a little bridge cir-
cuit padding here and there, get it
as close a: 0-2.

Getting the best out

Because of the circuit design, the
unit will perfor- best when the

"Coarse Amplitude" control is in
its highest position, pernittlng
appreciable attenuatitm with the
Fine" control. You will get mini-
nam noise products uhmever you can
use the "Tens" frequency switch,
also. The nature of the distortion
products is now mch mre clearly
defined and al-ost pure second hers
mic--idunl (or auplifinr measure-
M15.

So far, I have said nothing about
the square wave section. I do not
need to. Heath have undersold them-
selves here, because it is way, way
better than their mdcst claia of a
50 nanosecond rise tin. l checked
this on a Tektronix oscilloscope
which has an inherent rise time on
its vertical amplifiers of 12 nano-
seconds. W 'scope pictuw (see
Fig. 7) shows it is at least equal,
if not very such better than this--
so absolutely no changes are neces-
sary.

Not! I wonder if Heath might be
thinking of updating their Harmonic
Distortion Analyzer?

(while we have no Hulda informa-
tion from Benton Harbor, we do have
a intention of Heathl [lb-58 In
the work: for publication swan.
Tape bu!!! v! be particularly

in three had. -Zd.]