CSR-5 and CSR-5A Receivers


The RCA AR-88 receiver had been approved by the  Inter-Service Committee on Design as a general purpose communications receiver in WWII but  that did not stop the Canadian Navy from placing orders for the Marconi CSR 5 receiver instead of the AR 88.  Between March and August, 1943, this Service ordered a total of 740 sets.  New Zealand also placed an order for 100 of the sets. At the end of August, 1943, No CSR 5 receivers had been delivered, although original forecasts were set for July.  The latest forecast for start of deliveries was rescheduled to December, 1943. Many CSR's were placed into monitoring service at Special Wireless Stations which intercepted German U-boat traffic. Special Wireless Station Coverdale near Moncton was one such location.  Post war production quantity is not known at this time.

Some additional information on the CSR5 can be found on pages 49 and 50 of the Signals Production Branch document dated October 15, 1943.

coverdale_wrens_wwii.jpgWomen's Royal Canadian Naval Service (WRCNS, pronounced WRENS) operators are using Marconi CSR5 receivers at Special Wireless Station Coverdale in 1944. The receivers were manned in three shifts of 8 hours each. No one liked the graveyard shift or the ersatz coffee which helped  to keep everyone  awake. (National Archives Canada photo # PA 204141)
In peacetime and aboard Her Majesty's Canadian ships, many CSR-5's spent their working life receiving the Fleet Broadcast sent in Morse Code at 25 wpm or guarding the International or marine distress frequencies. Each receiver was connected to its own wall mounted speaker, but headphones were the order of the day. Loudspeakers were used when one Radioman had to guard more than one frequency. This was known as a loudspeaker watch.
MASS PRODUCTION: Fifty-three men could be accommodated at one time in the instructional receiving room, used for training basic classes in typing and Morse at the RCN Communications School at HMCS Cornwallis, Nova Scotia. At a later date, it was planned to expand to a Morse pool, with all exercises run from a central control room. Where else can one see so many CSR5's all at once? Cornwallis is now closed.  (From Crowsnest, May 1952. RCN photo DB-1301-1)

csr5_ haida.jpg
Serial 466 is one example on display aboard HMCS HAIDA. By 1969, the CSR-5 series was considered obsolete and was taken out of service in the RCN. Other than cleaning, no effort was made to restore the front panel of this receiver because it shows what controls were used the most while the set was in service. (Photo by Jerry Proc)


First designed by Canadian Marconi in 1942, this general coverage receiver was capable of receiving AM and CW signals in the low and high frequency bands Accuracy was 0.5 % at any frequency.


The band colours that the manual refers to are those observed in the RANGE window.  Band A on the dial, although designated as Mauve, is light gray in appearance on the dial itself in some receivers.

Band A - MAUVE 14.9 to 30.3 MHz
Band B -   RED  6.8 to 16/1 MHz
Band C -   GREEN  3.55 to 7.65 MHz
Band D -  ORANGE  1.5 to 3.5 MHz
Band E -  BLUE 195 to 518 KHz
Band F -  BROWN 79 to 207 KHz 

In some receivers, the band colours for Bands A and B are reversed depending whether the dial has a single or dual line pattern. This is an example of a dual line pattern. (Photo by Jerry Proc)
For whatever reason, when  Band A in the RANGE window is MAUVE, it does not have the corresponding colour on the dial.   Instead of mauve marking on the dial, it is light gray. Other variations of dial colours/markings are noted in the table below.
Single-line band marking, where the colors are: The dual-line dial pattern where the colors are:
Band A: Red
Band B: Gray
Band C: Green
Band D: Yellow/orange
Band E: Blue
Band F: Brown
Band A: Gray
Band B: Red
Band C: Green 
Band D: Orange 
Band E: Blue
Band F: Brown
Generally the receiver dial follows the colour scheme in the above table , howewer CSR5A S/N 842 is a single line track dial with Band 'A' being gray and 'B' being red.
This photo underscores the colour mismatch between the Band Indicator 'A'  in the RANGE window and the dial itself. (Photo by Jerry Proc)

After examining nine receivers, finally one is found with a real mauve marking for band 'B'. (Photo by Jerry Proc)
The single-line half moon pattern was found on CSR-5Y s/n112 and CSR-5A s/n 712 (RCN reconditioned 2-64) and originally on CSR-5A s/n 904 (with one with the peeled-off dial, below), so it doesn't appear to be related to serial number.

The RCN made a custom modification to their receivers, The "F" band (79 to 207 KHz) was adjusted 10 KHz low to enable the reception of the broadcast frequency of 73.6 KHz. This frequency is still assigned to Maritime Command as of 1994.

In marine service, one of the noted quirks of the CSR 5A was the habit of going off frequency in rough weather. If a large wave hit the ship, it would overcome the friction of the tuning gear assembly and knock the dial off frequency. There were no such things as frequency synthesizers or phase locked loops in those days.

The receiver features a 4 position crystal filter and noise limiter. To protect the receiver and especially in shipborne applications, there is a gas discharge tube between the antenna terminal and ground. Strong wind, moving across a ship's wire antenna, will eventually generate a rather large static charge which can damage a receiver's front end. The gas tube flashes over once the voltage level reaches 100 volts.

S/N 103, painted in black, has the "raised" Marconi  logo which seems to have been replaced with stencilling in later production runs. (E-bay photo) 
The intent of this photo is to illustrate the block lettering on on the front panel bezel (Photo by Rob Moffatt)

* Designators: Marconi CSR-5A  #110-930A .  RCN Ref 3A 107-1
* Antenna Input Impedance: 70 to 500 ohms.
* Frequency range: 79 KHz to 30 MHz in 6 bands excluding the broadcast band.
* IF Frequency: 575  KHz
* Frequency control: Variable oscillator or crystal.
* RF Gain: Step type control. 4 db attenuation per step at low end and 20 db per step at high end.
* Crystal filter: 4 position. Bandwidth info not provided in manual.
* Panoramic Display : Supports panoramic displaay via internal connection.
* Audio Output: 4 watts. Supports 10,000 ohm speaker with transformer input  or 500 ohm line or high
   impedance headphones.
* Case dimensions: 20.25 in wide x 10.5 in high x 15.25 in deep
* Weight : 68 pounds with case; 58 pounds when rack mounted  (power supply weight not included)
* Delivery start date: post August 1943 (?).
* Production: Mostly in 1944
* Applicable manual BRCN 2767 for CSR5A Type 110-930A
* Shockmounts: Barry Corp., Watertown, Mass. p/n C2080-5


There are six  CSR-5 receiver type numbers that we are aware of:

105865 CSR-5Y
110480Z CSR-5 receiver used in CM-ll transmitter-receiver
110480AZ CSR-5A receiver used in CM-l1A transmitter-receiver
110835W CSR-5 receiver 19 inch rack mount with dust cover (no cabinet) - all known examples are painted wrinkle black
110930Z CSR-5 stand-alone receiver in desktop cabinet
110930AZ CSR-5A stand-alone receiver in desktop cabinet

Type numbers 110480Z and 110480AZ were serial numbered sequentially, The changeover from
CSR-5 to CSR-5A came somewhere between serial number 350 and 384. The highest serial
number I have for this type is 399. The same holds true for type 110930Z and 110930AZ where
the changeover came at or after serial number 816 and the highest serial number located thus
far is 1070.


The CSR5 receiver supports a connection to a panoramic display, however Canadian Marconi did not suggest any specific panoramic type. Since the CSR5's IF is 575 KHZ, it may be difficult to locate a display  which is capable of displaying that IF frequency.  One approach might be to use a panoramic display that operates at the more popular 455 KHZ IF, It is suggested that a broadcast band (BCB) receiver be used as an IF  "down-converter"  which will shift the IF frequency from 575 KHz  to 455 KHz.. Using a  BCB receiver with a 455 KHZ IF, connect or couple the panoramic display output on the CSR5 receiver to the antenna  input of the BCB receiver. Then tune the BCB receiver to 575 KHz. Next,  couple the 455 KHZ BCB IF to the input of the panoramic display.  This is technique is unproven. It's presented here as a concept only. .A block diagram of the down-converter setup can be found here.

It is believed that the only possible application for a panoramic display would have been for CSR5 receivers installed at Canada's former signals intelligence stations.  However, no evidence  has ever surfaced to support that theory. There would be little,  if any, use at  any shore station.

CSR-5 vs CSR-5A

One commonly asked question is the difference between the CSR-5 and CSR-5A.

Radio collector Tom Brent comments " I have seen so many modifications and "hatchet-jobs" to these radios which only offer confusing clues but there are few items on the comparison list that unmistakably identify a CSR-5 from a CSR-5A. As originally manufactured, it is impossible to put a CSR-5 in a 5A cabinet and vice versa but I have seen numerous -5A cabinets with locator pins cut out to enable a -5 to be installed as well as -5 cabinets with the power connector opening enlarged to allow installation of the -5A receiver. I have seen crystal filter covers (which bear the part number) that have obviously been switched, audio transformers changed and many dial escutcheons with labeling that doesn't jive with what is behind the panel.

The escutcheon is a leading source of confusion as to whether a receiver is a -5 or a -5A. In the 1950's and 1960's the escutcheon was repainted and silk screened, part of a refurbishing program that many CSR-5's and CSR-5A's went through.

Some escutcheons were simply painted black and some were painted black and silkscreened or stencilled with the CSR-5A designation. The is no evidence so far of any repainted dial escutcheon with "CSR-5" (non "A") script on it. So, to set the scene here - At an RCN overhaul depot with perhaps dozens of radios being refurbished,  putting a CSR-5A dial escutcheon on a CSR-5 receiver could easily have happened and would have no operational or technical consequence. What they didn't consider was that 50+ years later, someone who would lovingly restore the receiver might be confused as to what model they actually had.

A typical Navy "reconditioned" sticker from 1963. (Photo by David Noon)
A repair tag from  March 1969 (Photo by Jerry Proc)
The 6SK7 vs 6SG7 tube issue still remains, at least in my mind, somewhat confused. There is the clear notation in the CSR-5 manual that states (contrary to what is shown elsewhere in the manual) 6SG7's have been substituted for V2 and V4 during the production run. Further, most of the CSR-5's I have encountered have had the original "6SK7" labels removed or scratched out and, in almost all cases, neatly relabeled "6SG7" with a rubber stamp. This could possibly indicate it was done at the factory although I suppose the repair depots could also have been issued with a rubber stamp. Curiously however, we still find this haphazard relabeling on the CSR-5A. I would have thought that if they could change the silk screen used to print the dial escutcheon and panel labeling, surely they would also change the one used to label the chassis. But who knows, maybe in the flurry of wartime activity at Canadian Marconi someone forgot to institute the change and the chassis was still being produced with the "6SK7" label at V2 and V4.

However, there are a few items that are almost impossible to change and provide a good method of establishing the actual model type:

1. The locations of the power switch and selectivity control are centered 1 1/2 inches below the top of the panel on a CSR-5 and 2 inches below the top of the panel on a CSR-5A.

2. A CSR-5A has 3/8" holes in the top corners of the back wall of the chassis (to accept locator pins in the cabinet); CSR-5's have no such holes.

3. A CSR-5 has a 2-screw terminal strip adjacent to the audio transformer; the CSR-5A version has a 3 terminal strip.

4. A CSR-5A has adjustable slugs for band E and F RF coils that are mounted on an add-on plate affixed to the top of the RF chassis (back-left corner); CSR-5's have no adjustment for these coils".

1 Crystal filter assembly is CMC part # 106-820. Selectivity control knob offset from shaft or rotary switch.  Crystal filter assembly is CMC part #106-852. Selectivity control knob in line with shaft of rotary switch.
2 Audio output transformer is CMC part #97690 Audio output transformer is CMC part #97690A
3 V2 (2nd RF) and  V4 (1st IF) type are 6SK7 – See note 1 below table  V2 (2nd RF) and V4 (1st IF) type are 6SG7 – See note 1 below table.
4 C16, C39, C57, C69 rated at 200 volts. C16, C39, C57, C69 rated at 400 volts.
5 C84, C92, C116, C118 rated at  200 volts. C84, C92, C116, C118 rated at 300 volts.
6 C98 – 26pf
C108 – 25pf
C98 – 30pf
C108 – 21pf
7 L12 CMC part  #106-586.  See note 2.  L13 CMC part #106-587 L12 CMC part #106-692. See note 2.
L13 CMC part #106-693
8  C133 and C134 C133 added to coil L28 (OSC. “C”)
C134 added to coil L29 (OSC. “D”)
9 R22: 70K OR 25K
R27: 400 OR 500
R34: 400 OR 500
R22: 25K?
R27: 100 TO 400
R34: 400 ohm
10   R62: 10K detector damping resistor added to  L20 (1ST DET.  BAND “A”) 
11 Front panel  Amphenol crystal socket only fits one size (type BY) of crystal.  Front panel  Amphenol crystal socket fits two sizes of crystals on part of the production run.
12 R58 load matching resistor on J2 is permanently grounded. R58 connected to link for grounded/ungrounded option. 
13 Power connectors are perpendicular to rear of chassis (oriented to rear). Power connectors are parallel with rear of chassis. One is oriented up and one oriented right.
14 Lines on dial for each band are single solid line. Band “A” line is RED. Band “B” line is MAUVE. Lines on dial are doubled (like a railway track).  Band “A” line is MAUVE. Band “B” line is RED.
15 Cabinet has vent holes in top, sides and back. Cabinet has vent holes in back and side only. 
16 Two L-shaped brackets, attached to back of cabinet hold down top of radio chassis. See Note 3. Two locator pins welded to the rear of cabinet engage holes in rear wall or radio chassis.
17 Single finger pull hole in lid of cabinet. Two finger pull holes in lid of cabinet. 
18 The audio output transformer 500 ohm winding has a two screw terminal board to optionally ground the center tap.  The audio output transformer has a three screw terminal board, which also provides a 250 ohm tap option.
19 C123 (2,000 mmfd., 500 volts) failing in service. With few exceptions, receivers up to serial number 816 have this condenser incorrectly connected from terminal 4 to terminal 6 of the output transformer T4, as shown in Canadian Marconi wiring diagram 111-900. C123 is correctly connected from terminal 4 to terminal 5 of transformer T4 in receivers S/N 816 and up. Also see note 4 below table.
Comparison table courtesy Tom Brent with additions by Meir Ben-Dror and Jerry Proc
Note 1:  V2 & V4 are shown on the CSR-5 parts list as type 6SK7 and on the CSR-5A parts list as type 6SG7. On all CSR-5 and CSR-5A receivers viewed so far, the tube position and type numbers were originally silkscreen printed onto the chassis. In all but one example of the radios seen so far, it appears that the original number (6SK7) at the V2 & V4 position has been removed and a new number (6SG7) added with a rubber stamp using black ink. The one exception is a CSR-5 (serial # unknown) which retains the original silk-screened “6SK7” at the V2 and V4 position. Of all the changes, this is the most significant one because it affects the fundamental operation of the receiver. Here is the official notification taken from the manual:
" In order to provide increased maximum sensitivity, the 6SK7 valves V2 and V4 have been replaced by the 6SG7 valves. This change has resulted in an improved AVC characteristic over the range from 10. to 100,000 microvolts. So as to exploit this feature, certain changes have been made to the circuit and these are shown on the schematic diagram".
Note 2:  L12 & L13 are the RF input coils for the two low frequency bands, “E” & “F”. They are not adjustable on the CSR-5 and thus adjustment screws do not extend through the top of the RF chassis. On the CSR-5A, these coils are repositioned and mounted on a small plate attached to left-rear corner of the RF chassis. They are adjusted in similar fashion (from a position above) to the other bands. This provides one of the easiest ways to recognize a CSR-5A.

Note 3: A CSR-5 receiver cannot be installed in a CSR-5A cabinet because the locator pins protruding from the back wall of the later cabinet prevent the earlier receiver from sliding all the way in. Conversely, a CSR-5A receiver will not fit in a CSR-5 cabinet because the power supply connectors require a larger opening to clear the back wall of the cabinet.

Note 4 : RCN Naval Serice (NS) Order 1008-78-1 dated Sept 9, 1944 requests (in part) that "Each receiver in service, or being placed in service, with serial number below 816 is to be checked. Condenser C123 is to be connected from terminal 4 to terminal 5 of transformer T4".

From the available data, the following patterns are in evidence. CSR5's saw three forms of service:

1) Receivers for rackmount operation. These would be supplied with the WE-11 (AC only) rack mount power supply and a protective cover. (It's too early to conclude if all rackmounted CSR5's and including the 'Y' variant came in black).

2) Receivers for standalone operation. These would be fitted in an vented enclosure and come with the standalone VP-3 AC/DC power supply.

3) Receivers destined for CM11's transmitters. These of course, would not come with any enclosure or either of the above power supplies.

Layered over top of that would be the phasing-in of the CSR5A model at some point in the production cycle and the various colour schemes discussed elsewhere in this document.


A big irritant in this receiver is the fact that many units do not have a serial number plate affixed to the left side of the chassis. It appears that serial number namplates were affixed to the top of the case and not the receiver iteself. Therefore, in a repair facility with multi CSR5's to repair, it would be in the realm of possibility that receivers and cases might get mixed up if the technicians were not careful.

Radio collector Tom Brent speculates on at what serial number serial number the model '5A was introduced. " I have CSR-5 serial numbers as high as 671 and CSR-5A numbers as low as 399. This indicates 2 possibilities, the first being that CSR-5 serial numbers started at 1 and went up to 671 (the highest number I have found so far) and possibly beyond. The second part of this scenario has CSR-5A numbers also starting at 1 and going up to 973 (the highest CSR-5A that I have found so far). Let’s round the highest numbers off a little and say that they produced 700 CSR-5’s and 1000 CSR-5A’s. Does it seem logical that the navy would have ordered that many in total? Does it seem reasonable that Canadian Marconi could have produced that many?

I believe there is a second possibility for the serial number question. Canadian Marconi could have employed a numbering system whereby the receivers shipped in cabinets were one series and the receivers included as a part of CM-11’s were another series. The information supplied by the numbers I have tabulated so far does nothing to disprove this theory and there is even some information to back it up but I need data from more radios before I can reach a conclusion".

As for the production run, serial 103 is the lowest one logged so far with Serial 1024 being the highest in the range. Any other contributions are most welcome. Contact jerry.proc@sympatico.ca . Please provide the following information:

1.    Where is the I.D. plate located? ie Cabinet or chassis?
2.    Serial number ?
3.    Type number (the WHOLE number)
4.    Specification number (sometimes missing)
5.    Pattern number (usually missing)
6.    On some early sets, the information is rubber-stamped with black ink onto the side panels (close to where they meet the front panel)
7.    Colour and finish is the front panel?

To add another element of confusion, some CSR-5A’s have an I.D. plate that identifies them as a CSR-5!  Some CSR-5 receivers also have a dial escutcheon that says "CSR-5A". It is possible that during refinishing (in the 1950's???) some dial escutcheons were silk screened with the script "CSR-5A" and then slapped on whichever receiver was being refurbished(?).

Webmaster's note: As of March 2014, the higest CSR5A serial number is 1062.


VP3 - AC/DC Supply # 110-540  RCN Ref 3D/101

This supply provides the following voltages to the receiver:
250VDC at 110 ma and 12.6 VAC at 3 amps.

The VP3 power supply for the CSR 5A was designed to operate from 115/230 volt 25/60 Hz AC power or 12 VDC. When switch over to DC power, Marconi designed two interlocks to ensure that no damage could be caused by inadvertent operation on the wrong power source. To switch from AC to DC operation, a five pin interlock plug had to be moved from one socket to another. Subsequently, the AC line cord had to be disconnected from the wall socket and inserted into a special chassis mounted receptacle. One of the prongs caused a switch to open up. VP3 power supplies also acquired a reputation for fusing the contacts on the vibrator thus frying the primary winding on the power transformer.

The VP-3 power supply came in two versions. One version covers S/N 101 to S/N 1200 while the other covers 1200 and up. Two major differences are observed when the schematics are compared. The part number of choke L4 is different.  For the older serial number range,  the part number was 116-277 while the newer range is #97704. Also, the position of F2 was moved.  In the newer series, it now protects the AC input directly instead of being in series with F1 in the DC input line.

VP-3 with cover off. Note the mauve bakelite plug which is one of three interlocks for AC or DC operation. When used with DC, the AC plug must be removed from the wall receptacle and inserted into the socket to the left of the rectifier tube. Then the bakelite plug must be moved from the AC to the DC socket. Lastly, a toggle switch (locked with a metal bracket) must be flipped to the DC position.  (Photo by Greg Farrell)
VP-3 bottom view. Note the copper-clad matte finish chassis. This VP-3 came with a service tag indicating that the unit had been installed in HMCS Fundy 159 in January 1964. (Photo by Greg Farrell)
This is how the spare vibrator fits into the cover. (Photo by Greg Farrell)

The Marconi VP-3 power supply is at the right and a near replica built by Jerry Proc VE3FAB on the left. The VP-3 It is not wearing its original deteriorated light gray crackle paint finish since it had been repainted with Tremco Grey Rustclad paint. (Photo by Jerry Proc)
Another view of the VP-3. This example is missing the vibrator which plugs into the 4 pin socket. (Ebay photo)
This is a CSR5 power test cable with inline measurement capability. The 12 VAC  is monitored with an incandescent lamp while  the 250 VDC line is monitored by a neon bulb. Measurement points are conveniently provided on the front.  The male and female connectors are Cinch-Jones P-303 and S-303 respectively. (Photo by Jerry Proc)
WE-11 AC Only Supply  #110-973

The WE-11 power supply was used in applications where only 115/230 VAC  60 Hz power was available. It was available in either a rackmount version ( 32 pounds)  or a enclosed bench type unit (30 pounds).

The front panel of the WE-11 AC power supply is extremely simple bearing only a power switch and pilot light. (Image courtesy Canadian Marconi)
An actual WE11 painted with black crackle paint. (Photo by Jason Ingraham VE1PYE)
WE-11 bottom view. (Image courtesy Canadian Marconi)
WE11 nameplate. (Photo by Jason Ingraham VE1PYE)

Not much remains that hasn't been replaced on this WE11.  The two (slightly oversize) transformers are replacements, with only the filter choke being original. The wires and line cord have also been replaced since they were deteriorated and unsafe. A fuse holder, very evident here,  was added to the front panel by a previous owner.  This picture does however, illustrate the general construction of the WE11.  (Photo by Jason Ingraham VE1PYE)
Underside view of the WE11 chassis. Not all the parts are original.  (Photo by Jason Ingraham VE1PYE)
For anyone wishing to construct their own CSR-5 power supply here are the receiver's requirements:
B+ 250 VDC at 110 ma;  Filaments 12.5 VAC or VDC at 3 amps. Suggested schematic available at the bottom of this document.
csr5a_power_connector1.jpg csr5a_power_connector2.jpg
This is the common power connector housing for a cabinet-mounted CSR5A. Receivers  in standalone cabinets can also use the connector housing shown in the rightmost photo.  (Photo by Jim Brewer)  Power connector housing for CSR5A fitted into a CM11. Receivers used in CM11 cabinets can also have the connector housing shown in the leftmost photo. (Photo by Jerry Proc)
Standalone Speaker #110-823

This is an 8 inch permanent magnet speaker mounted in a case 12' wide x 12 wide x 6 in deep. A two screw terminal strip is provided for the connection to the receiver.

csr5_speaker_front_wf2u.jpg csr5_speaker_rear_wf2u.jpg
Front view of #110-823 speaker with original gray crackle paint. It comes with mounting brackets for attachment to a ship's bulkhead. (Photo by Meir Ben-Dror, WF2U) Rear view of another #110-823 speaker showing the high impedance transformer and original, green hammertone paint peeking around the back perimeter. (Photo by Meir Ben-Dror, WF2U)
csr5_speaker_blk.jpg csr5_speaker_ltgre
This #110-823 speaker is painted in black crackle. (Photo by Jason Ingraham VE1PYE) This variant of the speaker uses a black grille screen and bezel with a light grey finish on the cabinet.
csr5_speaker_screen1.jpg csr5_speaker_screen2.jpg
 This #110-823 speaker is fitted with a screen instead of cloth behind the grille. (Photo by Greg Farrell) Rear view. Speaker p/n is 116-260 wile Transformer bears p/n 4D126G72 (Photo by Greg Farrell)
Another variation in the grille material  
Rackmount Speaker #110-836

The panel is 19 inches wide by 10.75 inches high

#110-836 speaker panel .The manual does not say whether these are two speakers wired in parallel or whether they are two independent speakers. (Image courtesy Canadian Marconi)


For operation using crystal control, the crystal must be 575 KHz higher than the desired receiving frequency when the crystal is operating on its fundamental frequency. RCN operators also hand marked CSR-5 crystals with the frequencies of the second harmonic and the third harmonic. The receiver could be tuned to operate on the fundamental or the second or third harmonic.  The following standard RCN crystal frequencies were available for the CSR5 receiver.

The CSR5 or CSR5A can accept a  type 'BY' crystal which is depicted on the right,  while the CSR5A (only)  can accept either of the crystals depicted in this photo. The case style of the leftmost crystal is not known at this time, however its dimensions are 1.125 in  wide by 1.25 in deep by 3/8 in high. The pins are on 1/2 inch  centres. (Photo by Jerry Proc)


During an alignment on the IF stage, one troubleshooter noticed that the top iron core of T1 in the IF crystal filter was quite broad and had no noticeable peak anywhere in its travel.  On removing the
filter module, he noticed a "clunk" as it was moved around. This was traced to the 575 KHz filter crystal. Opening the crystal and fearing the worst, he discovered the center element of the crystal  was free to slide around in the holder and was only guided by supports on either side. This turns out to be normal since the crystal in the filter circuit is of the "loose resonator" construction (exact name is unknown) versus the "fixed"  type whose quartz element is fixed between two, spring loaded, metal electrodes. Loose resonator crystals work on the principle of a quartz element operating in a small air gap and free to slide back and forth, however the exact principle is not known at this time. In this instance, the disassembly and reassembly of the crystal cleared the problem. So why did Canadian Marconi choose to use a loose resonator crystal (575 kHz) for the filter circuit and fixed resonator crystals for fixed frequency operation of the receiver?

Exposed: The two pieces which constitute the guides are ground glass and are slightly thicker than the resonator. The resonator element is allowed to move freely within a .003 inch air gap and from side to side. (Photo by Jason Ingraham VE1PYE)
This 575 kHz filter crystal, with identical Canadian Marconi identification to the one above, was pulled from another CSR-5. With its three glass support elements, why does this one differ?  (Photo by Jim Brewer) 
575 KHz crystal nameplate. (Photo by Jim Brewer) 
In the September 1934 edition of QST magazine, James Millen, President of the National Radio Company explained these "loose resonator" crystals. Here are key extracts from that article.
"Perhaps the greatest misunderstanding centres around the resonator crystal and its holder. It is surprising how how many amateurs do not realize that the resonator crystal must NOT be under pressure. For resonator crystals, we have found that an air gap of approximately .003 inches to be essential. In order to properly maintain this air gap, we have found it most practical to separate the holder plates by means of two carefully ground glass parallel bars or spacers. The crystal element itself is placed between the spacers. From our correspondence many amateurs believe we are using three "trick" crystals apparently concluding that the spacer bars are made from quartz.

In an earlier model of resonator crystal, we used a bakelite spacer ring surrounding the crystal. This ring was carefully ground so as to provide the .003 inch air gap between the crystal proper and the plates. It was startling as to how many amateur complained to us that upon opening their crystal holder they were surprised to find that we had been careless in using a spacer that was thicker than the crystal and consequently prevented the holder plates from touching the crystal. They were filing down the bakelite spacer ring until it was thinner than the crystal.

Then there is the matter of polarity of the crystal holder. When using a holder with horizontal plates, it is important that the holder be inserted in its socket the same way at all times otherwise it will be necessary to rebalance the bridge circuit".

 Crystal filter compartment with cover off. (Photo by Jerry Proc) 



CSR5 receivers came in five known paint schemes. When standalone, in their own cabinet, the paint finish was of the crackle type. When part of the CM11, the finish was typically glossy, light gray.

1)  When the CSR5 was part of a CM11, it had white lettering over top the light gray, smooth paint.

This CSR5 is part of the CM11 in Radio 2 aboard HMCS HAIDA. (Photo by Jerry Proc)
2)   This pale green was a very common finish for the CSR5. An exact paint match was made from the CIL Paint line in 1993. It's called Silver Lace #10GY49/081-7 but its availability is unconfirmed in 2007.  Note the transparent type decal front panel markings, It is suspected that these were applied after the receiver left the factory, perhaps as a result of repainting the front panel. In cases such as   this, the dial bezel will lack the stylized "Marconi Canada" script.
This CSR5 S/N 122 is from the collection of Richard Brisson. (Photo by Richard Brisson)
3) Light gray crackle paint with black, silk screen lettering as below .
csr5a_light_ gray.jpg
This example was painted with a beige/grey hammertone finish.  Note the "dual semicircles" for each band marking on the dial.  (Photo by Jerry Proc)
4) Medium blue-gray with white lettering.
This example, painted dark blue,  is found aboard HMCS HAIDA. (Photo by Jerry Proc)
5) Black crackle with white, silk-screen lettering as in the case of this CSR-5
This CSR5 S/N 219 belongs to Jason Ingraham.  (Photo by Jason Ingraham VE1PYE)
6) Black crackle with white, silk-screen lettering as in the case of this CSR-5Y.
This CSR-5Y example is owned by Meir Ben Dror. (Photo by Meir Ben-Dror, WF2U)

This CSR-5A, S/N 103, is painted in black "alligator skin" finish with stencilled lettering on the front panel. It is believed that this is either a pre-production or prototype unit. Band 'A" has a real mauve dial marking.  (Photo by Greg Farrell). 

S/N 103 has also been fitted with this raised chrome logo perhaps suggesting it may have been a protoptype and the logo was borrowed from a  production radio. (Photo by Greg Farrell). 
7) Here are two examples of restoration no-nos. This is how a CSR5 should never be refinished - aluminum paint with Dymo tape lettering.
Many decades ago it was believed that the supply of surplus gear would last forever, so modifications like these were typical of the period. Oddly enough the small, black engraved plate indicates Serial number 1 but that has to be taken with some doubt since the three pin crystal socket suggests that it's a CSR-5A. It also looks like the front panel lettering composed of name tags which have been removed and replaced with DymoTtape labels.  (Photo source unknown)
This example also illustrates an improperly restored CSR5 receiver. Work such as this significantly reduces the value of the artifact and makes it undesirable to serious collectors.
In the lower photo someone has added an inboard power supply to the cabinet lid. The new owner will be stripping out these components and constructing an outboard supply. (Photo by Carl Nord WA1KPD)
8) Was This a Pre Prototype Receiver?
This might be a pre-protype CSR5 receiver. The bands on the dial are monochrome; the crystal socket and crystal in/out switch are not there and it appears a meter was added by perhaps someone who owned the receiver after it became a surplus item. The cabinet paint is kind of a dark green bordering on black. This example is held by the C&E Museum in Kingston. (Photo by Jerry Proc)

The CSR5 cases were manufactured in to variations.

CSR-5 cabinets have 2 ventilation screens on the lid, 2 ventilation screens in the back and 1 ventilation screen on each side.

CSR-5A cabinets have 2 ventilation screens in the back and 4 ventilation screens on each side (none in the lid).

CSR-5 cabinets have a rectangular hole at the lower right-rear for the power connections. Both connections are oriented perpendicular to the back wall of the chassis. In other words, they go straight out the back from the little box they are mounted on.

CSR-5A cabinets have an irregularly shaped hole for the power connection because the cables connect to the top and side of the small box mounted on the rear of the chassis. CSR-5A cabinets also have 2 steel pins mounted on the inside of the back wall of the cabinet that mate with holes on the left and right side of the back wall of the receiver chassis.

A CSR-5 receiver will not mount in a CSR-5A cabinet unless the pins are removed.

Conversely, a CSR-5A receiver will not fit in a CSR-5 cabinet unless the rectangular hole is cut out to allow the Cinch-Jones type power connectors to pass through the back wall of the cabinet.

The final difference between the two cabinets is that CSR-5 cabinets have one finger pull and CSR-5A cabinets have 2.


Meir Ben-Dror, WF2U had a badly chipped half-moon dial which he repaired.  Here was a summary of the process he used to make a new one. "First I made a high resolution scan of the dial.  The paint was missing in a few little sections in the grooves of the orange and green  band/frequency markings so I used Photoshop to touch up the markings I printed out the dial on heavy, semi-gloss photo paper, and coated it with Krylon protective spray. The reproduced dial looks indistinguishable from the original. It even looks 3 dimensional due to the high resolution scan and some enhancements I did with Photoshop. Then I matched some acrylic artist's paint and touched up the markings on the original dial then wiped off the excess from the surface. It is not known at this time if the heat from the dial lamps will have any adverse effect on the new dial indicator".

Dials came in two styles. One style had the single half moon markings for each band while the other style had the double half moon markings which resemble a railroad track. These differences are shown in the two photos below.

This dial, from a CM11 mounted CSR5 s/n 392, was in a deteriorated state. The dial pointer is held on with a single Bristol screw and is easily removed. Note that the colours of A and B bands are reversed to that of the dial which was scanned from a CSR5A.  (Photo by Meir Ben-Dror, WF2U)
This is the finished reproduction '5A' dial with pointer reinstalled. (Photo by Meir Ben-Dror, WF2U)
The final product - a professional looking dial .(Photo by Meir Ben-Dror, WF2U)

Ever wonder how a CSR5 dial looks with the silk screening removed? Here is an example. (Photo by Jerry Proc)

In the CSR 5A, the band-switch assembly has been wired into a sub-chassis which can be detached from the main chassis. This operation should be never be attempted by the inexperienced. First, you extract the receiver from its case and detach the bottom cover plates - do not be concerned over the 30 screws that secure the plates. Next, desolder 29 connections as outlined in the manual. Following that, there are another 20 screws to remove in order to physically detach the RF turret. Do not go insane in the process, or you won't be able to get the pieces back together. This brief glimpse of 1942 radio maintenance has been presented for those who have complaints about current  manufacturing methodology!

Bottom view of chassis with cover plates removed. (Photo by Jerry Proc)
Top view of chassis. The RF turret is at the left side.  (Photo by Jerry Proc)
Rear view of chassis. Note the holder for the Allen keys at the right side of the photo. By extracting the chassis about 6 inches, an operator could access the Allen keys and  easily tighten a loose knob. The box with the octal plug is a home brew modification of connecting power to the receiver. (Photo by Jerry Proc)
There is an undocumented modification made to the CSR5 receivers specifically fitted in MCS HAIDA. 

The standard way to connect an antenna to the CSR5 is to use the terminal strip at the back of the set. The RCN did not do it that way. Instead,  they added a SO-239 connector to the right rear side of the chassis and also drilled a clearance hole into the cabinet to accommodate the SO-239 . The coax cables connecting the CSR5s to the aerial exchange board were hardwired using cable clamps and terminated  with a PL-259 connector at both ends. The centre pin of the SO-239 was connected in parallel with the antenna screw on the rear terminal strip.  So a cabinet which lacks the clearance hole will be incompatible with a cabinet that is modified with an SO-239 receptacle. 

So there is a possibility of finding CSR5s with or without a SO=239 at the back of the cabinet. The CSR5 in the CM11 did not have this modification.  It is not known if the SO-239 mod was done to other CSR5s aboard other ships. 

Output connection for Panoranic adapter and taps for audio output impedance. (Photo by Jim Brewer)
CSR5A top view of RF deck.(Photo by Greg Farrell).

When a CSR5 was intended for rack mounting,  it was provisioned with a protective cover such as this. This is the cover for S/N 219. Note the louvers with screen backing and the other differences in the cutouts when compared to the example below. (Photo by Jason Ingraham VE1PYE)
This cover from S/N 103 only has one large opening for ventilation which covered with a perforated screen. (Photo by Greg Farrell).

The dust cover attaches to female Dzuz latches which are mounted on the the chassis sides at the back. (Photo by Jason Ingraham VE1PYE)

S/N 103 has its serial number stamped in ink above the audio jacks but no nameplate. (E-bay photo) 
There may be an easy explanation for the rubber stamped serial number. This may be an early, "bid sample" or a prototype for the production line. Until a firm order was received,  it would not make financial sense to provide the tooling to produce a metal nameplate. That would come after a firm order was received. Perhaps Marconi simply did not have all the parts at hand as production started.


The following CSR5 technical information is available for download. The CSR5A schematic was redrawn by Denis Chouinard VE2DSH. David Noon provided the CSR5 erratum. In 1959 the RCN developed a test agenda for the CSR5 which has also been included.
Documents for CSR-5 
CSR-5 Manual   May 1944, 35 mb. Scanned by Gerry O'hara
CSR-5A Receiver Schematic (Redrawn by Denis Chouinard) 
CSR-5 Receiver Schematic - BMP format. Scales to 11 x 17 inch paper 

Other supporting documents:

CSR-5A Dial Assmbly Drawing
CSR-5A Accessories Listing
CSR-5A RCN Test Agenda
CSR-5A Speaker Arrangements
CSR-5 Erratum 
CSR-5 IF Tag Boards - Drawn by Gerry O'Hara

Home Brew Power Supply (Full Wave Rectifier Type) 
Drawn by: Gary I. Biasini 

Home Brew Power Supply (Bridge Rectifier Type)
Drawn by Jerry Proc

Gerry O'Hara has produced an extensive article on the restoration of CSR5 receiverrs


Based on the few snippets of information which have been located so far, a tentative time line for the CSR-5 is beginning to emerge.

*  We know that production of CSR-5 receivers in a stand-alone cabinet (Type 110930-Z) did not begin until late 1943, possibly early 1944. Further, we know that the switch from CSR-5 production to the CSR-5A model probably occurred in July or August 1944. A  photo  from this time frame confirms that the CSR-5 was being installed aboard new ships and the SMR-3 was no longer standard receiving equipment.

* September 1942: CSR-5 receivers in production as Type 110 480 installed in CM-11 transmitter receivers. (Evidence from notation in CM-11 manual, production likely started earlier)

* ???? 1943: Rack mount CSR-5 receivers (Type 110835W) in production. (Evidenced by WE-11 ID plate patent date of 1927-43.)

* March - August 1943: RCN places orders for 740 receivers, New Zealand places orders for 100. Presumably, these orders are for Type 110-930 (stand-alone receiver in cabinet).

* July 1943: Expected date for production to begin.

* August 1943: Production had not commenced; production now forecast to start December 1943 .

* September 1944:  Serial #816 is the last of the stand-alone receiver production.. At this rate of production, it would suggest that all of the CSR-5/5A's  found so far, up to the highest serial number of 1025, were built before the cessation of hostilities in 1945. ( Serial number 816 was stated in Naval Service Order 1008-78-1 dated Sept 2, 1944)

Is this the production line at Canadian Marconi circa 1944? No...its Meir Ben-Dror's workbench where a group of CSR-5 receivers are being restored back to their former glory.  (Photo by Meir Ben-Dror, WF2U)
This is the workbench of Jason Ingraham VE1PYE. CSR5 S/N 219 is undergoing troubleshooting. Behind it is the WE11 power supply and the Marconi speaker. (Photo by Jason Ingraham)

September 2007: Radio 1, starboard view aboard HMCS HAIDA in Hamilton , Ontario. All these receivers were brought back to working order by Jerry Proc. (Photo by Jerry Proc)

Contributions and Credits:

1) http://www.qsl.net/ve3bdb/signalshistory4.htm by Bob Cooke VE3BDB
2) Meir Ben-Dror, WF2U <wf2u(at)ws19ops.com>
3) Tom Brent <tgb(at)telus.net>
4) Denis Chouinard <denischouinard(at)enter-net.com>
5) Richard Brisson's Boatanchor web page:  http://home.ca.inter.net/~hagelin/boatanchors.html
6) CSR5A Manual #110-434A
7) Jim Brewer <snack.235@sympatico.ca>
8) Jason Ingraham <the_doctor31(at)hotmail.com>
9) Greg Farrell, Santa Rosa, California  <gregf(at)pon.net>
10) Carl Nord WA1KPD <chnord(at)comcast.net>
11) David Noon  <va3dn(at)execulink.com>
12) Gary I. Biasini <gary.biasini(at)shaw.ca>
13) Rob Moffatt <rmoffatt3(at)gmail.com>
14) History - Signals Production Branch , Department of Munitions and Supply, October 15, 1943
15) Gerry O'Hara [gerrycohara@gmail.com]

Back to Equipment Listing
Dec 14/11