Drake TR-7 Improvements
Marinos Markomanolakis, SV9DRU
January 10, 2017
View comments about this article!
"Editor's Note: Due to the popularity of some of eHam's older articles, many of which you may not have read, the eHam.net team has decided to rerun some of the best articles that we have received since eHam's inception. These articles will be reprinted to add to the quality of eHam's content and in a show of appreciation to the authors of these articles." This article was originally published on: 03/04/2011
Drake TR-7 Improvements
By Marinos Markomanolakis, SV9DRU
The TR-7 remains one of the top choices of a radio which offers extremely clean Rx performance,
robust construction and no infestation by the all too usual nowadays DSP sound “enhancement”
gimmicks which manufacturers are using to lure customers into buying new rigs.
This excellent rig can be further improved in a couple of areas, with a few straight forward
interventions that will bring it to the level of functionality that we have come to expect of the
modern radios. This topic has been covered a few times in the past, and I would like to share my
experience in two areas:
1) Installing the Cumbria Designs x-lock stabilizer in the TR-7
2) Improving the SSB TRx audio quality
Installing the Cumbria Designs frequency stabilizer, a new twist to an old project.
Having an analog PTO gives the TR-7 a low phase noise frequency generating ability, but also its
inherent, temperature depended frequency drift. This seems to stabilize quite a bit after 1 hour of
operation and can be reduced by installing the optional fan, but cannot be totally eliminated.
The Cumbria Designs X-lock stabilizer kit http://www.cumbriadesigns.co.uk/x-lock.htm is
offered at a very reasonable price and has several refinements over the original Huff-Puff
stabilizer as well as other similar circuits that can be found on the internet. Following a few
simple steps, it can be integrated in the TR-7 and provide rock solid frequency stability from a
Installation Overview: A picture is worth 1000 words!
First step: Where and how to install the X-lock in the TR-7:
As luck would have it, 2 of the 4 corner holes in the kit’s PCB align perfectly with 2 of the
supporting screws in the back left corner of the TR-7’s underside.
While someone could potentially try to locate longer screws and pass them through the mounting
holes of the X-lock, through short spacers and into their original positions, I choose to do it a
Electrical wire small binding posts were obtained from a local electronics store. After removing
their plastic insulation, I bent them 90 Deg (Photo 2) and adjusted their cylinder diameter (using a
flat fine file to remove some of their metal on top, followed by bending them into a smaller size
cylinder) in order to make them fit into the PCB mounting holes.
After installing them on the Tr-7 using the original hardware, I fitted the X-lock through them,
using a temporary spacer to keep a safe distance between the under-surface of the motherboard
and that of the stabilizer, and soldered them. The X-lock mounting holes are grounded and
through plated which makes for a very secure and mechanically stable connection.
The stabilizer PCB is rather small size and supporting it by its 2 side holes gives adequate
stability. A 3rd support structure at the opposite end was placed under the X-lock PCB as shown in the next picture, mainly to keep a safe distance from the TR-7 motherboard. Since the distance
between the pins of the motherboard under the X-lock is rather marginal, I cut the tips of these
pins flush to their solder joints and placed a layer of electrical tape over them.
Next, one must pay attention at the clearance of the higher points at the component side of the Xlock from the metal cover of the TR-7. These are the tops of the 100µF and 10µF caps, and even though they should not have any significant voltage difference with ground, I placed one layer of thin adhesive tape on the area of the bottom cover over the stabilizer.
Step Number 2: How to interface the X-lock to the TR-7 circuit:
Since this stabilizer includes its own reference crystal, the only connections needed to the TR-7
are the PTO sampling frequency (RF input), the power supply (13,8V) and the correcting voltage
to the already existent RIT circuit built in the TR-7 (X-lock DC output). This makes for an easy
and straight forward step #2.
The PTO connection is via a small caliber 50 Ohm coaxial cable RG-174 from the posts with the
PTO sign of the motherboard to the RF in of the X-lock. The +13,8V power supply comes from
pin# 9 (counting from the right when the unit is upside down) of the PS board assembly of the
As for the DC control voltage connections, I followed the instructions from the original article at
http://downloads.hanssummers.com/stbham.pdf You essentially need to unsolder the 2 RIT
cables (brown wires) from the motherboard and connect them to the X-lock DC output voltage
via a simple resistor and diode circuit as seen in the following schematic.
In my case there was not even a need to re-center the RIT after integrating the X-lock to the TR-
7, but this can easily be accomplished by following the instructions in the above article or the
service manual, should you need to do it.
The last and most enjoyable step is the installation of the stabilizer status indicator LED, which
fits perfectly in the place of the “FIXED” bulb indicator. The 2 LED control wires need to be
connected directly to the X-lock and the ground LED wire can be connected to the ground
connection of one of the nearby switches (UP DOWN FIXED). Here I made a slight modification
to allow the “FIXED” light to shine in its original RED color through its red filter by reversing
the red-green LED connections.
Normally the X-lock LED will shine red when the device is unlocked during fast turning the
PTO, and returns to green when the stabilizer has locked the PTO. The problem is that green light
will not get through the already existent red filter of the original TR-7 FIXED light assembly and
thus one can only see the LED in its “unlocked” red state. By reversing its connections though, it
will shine red, and be visible as “FIXED” when the stabilizer has locked the PTO frequency,
which I think is more appropriate. It was very nice indeed for Drake to include the “FIXED”
illuminated indication 30 years ago!
Finally, while the default delay time for the X-lock to “grab” and stabilize the PTO once any
movement of the tuning knob has stopped is 2 seconds, I found it more convenient to increase it
to 10 seconds (using the process described in its manual) for my tuning habits. I like to try and
bring the incoming station as close to its correct frequency as possible by small precise fine
tuning movements of the PTO knob, which most of the time take longer than 2 seconds.
This may result in me fighting with the X-lock while trying to fine tune. After expiration of its 2
second tolerance it will detect my fine tuning as slow drift and tries to correct it!! Thus both me
and the machine are getting aggravated… Everything can go back to normal though, if you
increase the stabilizer's off time, say to 10 – 15 seconds which would give it “a brake” while you
are trying to fine tune your new extra stable PTO…
Improving the TR-7 SSB audio quality
The stock SSB filter in the 2nd IF is 2,3Khz wide. While this will give good communication
quality audio, it leaves something to be desired, especially if you listen to it side by side with a rig
using a 2,7Khz filter like a TS-830S / 930S / 940S, an Atlas 210X etc.
I am not referring to the “ESSB” sound that some fellow operators like to use, but rather to a
“dryer quality” sound of the 2,3Khz audio which, without providing an appreciable improvement
of selectivity compared to the 2,7Khz one, removes some high frequency spectrum form the
audio which makes listening less rewarding and to some extent may even reduce the intelligibility
of voice especially when listening to lower signal strength, close to the noise floor of the band.
While searching for an alternative filter, I came across an offering from Sherwood Engineering at
a width of 2,85Khz. This is a little wider than the above mentioned 2,7Khz “optimum, but it does
seem to have excellent shape factor, which allows for richer sound reproduction without
becoming a “barn door” filter allowing excessive interference or “hiss” to spoil its performance.
Practically, for 90% of SSB work I have found even the 1,8Khz filter to offer little additional
selectivity or QRM suppression, with the exception of crowded band conditions.
Substituting the 2,3Khz original filter in the first position of the filter board of the TR-7 is straight
forward, with caution when de-soldering the original in that you have to use a 40W soldering
iron. Also I ended up prying it out gradually using a very small flat (jeweler's) screwdriver close
to each of its connections while heating its connecting posts. It is rather hard to try removing all
the solder from each connection since even the smallest amount of residual solder is enough to
prevent any movement of the filter from its tight connection posts through the double sided PCB.
This is the first and most important step in improving both the Rx and Tx audio bandwidth. I did
undertake one more step, which allowed slight further improvement of the Rx audio with little
additional work or risk.
Following the suggestions seen in the PA0CMU site
http://members.ziggo.nl/cmulder/drake.htm#bookmark6 I experimented with changing a few
capacitors and 2 transistors in the 2nd IF board, at the final audio amplifier circuit, in order to
improve its frequency response as described in the article.
After trying the proposed changes, I ended up implementing the following:
Changed Q1111 and Q1113 from the general purpose 2N3904 to the low noise BC549C model.
(Warning observe the different pinout!). And Changed C1116, C1164 and C1167 caps from 0.01µF to 4,7nF. This increased the high end Rx audio response and made a significant improvement in the intelligibility of marginal SSB signals, at least to my ears.
I did not change the other caps (actually reinstalled the originals after trying the proposed
changes) for 2 reasons: a) Increasing the values of C1150 and C1152 gave the audio rather too
much lows, which resulted in listening fatigue (again for my taste). b) Changing the C1173 to
220pF from the original 330pF, resulted in a constant low level audio “whine” mostly noticeable
when listening at low audio gain or with the antenna disconnected.
The final result after the above changes, gave my TR-7 a smooth a pleasant audio which rivals
that of my Kenwood and JRC radios, while maintaining the legendary Drake clean “tube like”
audio quality. All of them are easily reversible if one wants to go back to the original
configuration, and with the exception of the new crystal filter, will cost less than $3 in parts.
Admittedly, I did perform the changes in the audio amplifier circuit after having substituted the
filter with the 2,85Khz one, so can not comment on their isolated effect, but of course these 2
steps are independent from each other.
Hope you enjoy keeping an excellent classic as a TR-7 working and improving it as much as I do.
Marinos, sv9dru / ki4gin
This article has expired. No more comments may be added.
Drake TR-7 Improvements
by W3RSW on January 11, 2017
Mail this to a friend!
I see that a new version xlock3 is being made available since existing units are out of stock .
This is an excellent article and well worthwhile to put these mods in a TR-7.