Rebuilding the Moreton Cheyney Amplifier

The amplifier as you can see from original photographs was in a dire condition, having been stored in wet conditions. Is it worthwhile to attempt to tackle a rebuild? Well, this amplifier, which was designed to accompany the Moreton Cheyney receiver, is almost certainly, like the receiver itself, the only one in existence.

Whilst the receiver has only been slightly modified (or probably bodgingly repaired rather than modified), the amplifier has quite certainly been modified to a considerable extent. Rather than guess the original design, I've decided to rebuild it in it's last working state. Fortunately there are clues to help me to do this. For example, the amplifier last used a pair of push-pull KT66 beam tetrode valves driven by a pair of EF37A pentodes. The power supply takes up the lion's share of the chassis.. in fact parts have been added on a sub-chassis screwed to the rear. It seems there are probably three HT lines and several LT feeds with two cables running to the receiver. All of the wiring and cabling is completely decayed. Most of the soldering is in a very poor state and it's quite possible that one or more of the major wound components will prove to be unserviceable. In fact, if one of the critical parts is beyond redemption the whole rebuild project might be cancelled and any serviceable components consigned to the junkbox. Below is a set of pictures showing the current condition of the amplifier. Rust has been treated but little else has been done.

Original connections to the amplifier comprises two cables which plug into the receiver (power plus audio), mains and single loudspeaker.

For some reason the last owner decided to add an extra HT feed hence the sub-chassis carrying a choke and large smoothing condenser. Possibly he found the KT66 current draw affected the pre-amp output or the KT66s needed a reduced HT voltage for their screens? The latter is favoured because the original design perhaps used PX4 triodes with a high anode voltage.

The ceramic valveholders almost certainly replaced old B4 or B5 valve holders?

In view of the way the components are fitted (especially that bias pot) it looks like the owner was either unhappy with results or just abandoned the whole project before it was finished? It remains to be seen if one of the major parts is duff....

Below are the circuit diagrams of the amplifier and power supply with component identification.

The aim is to remove all the components and wiring from the chassis, test the parts and then if all are OK, or if I can easily source replacements, grind paint and rust from the chassis and sympathetically tidy up everything else.

As soon as I'd started I discovered it was impossible to shift many of the securing screws used to hold components in position. There are a mixture of nuts securing 4BA screws. About 60% of the screws sheared off or had to be drilled out due to their threads being rusty.

In fact dismantling the chassis was a horrendous job due mainly to the combination of locknuts and screws with very thin slots chosen by the Moreton Cheyney designers. To compound difficulties they also used very long screws with non-standard nuts which of course had rusted badly and were too long for a standard socket to be used. Below, a selection of pictures showing progress.

Above, the four block condensers, all of which measured nominally OK with my tester but will need further testing with a high voltage applied across their terminals.

Cleaning and further sandpapering revealed rather scruffy examples. A search in my collection of cut price bargain collection suggested a gloss paint in a shade of pale duck egg blue. I decided against preserving the lettering.

The finished articles which will go nicely with the repainted chassis, below.

I bought the chassis paint thinking it was black because the writing on the can said "colour as lid".. which was black. Gloss white is a zillion percent improvement over brown rust. A second anti-rust treatment under the chassis gave me a pretty good bare metal finish which I left unpainted for good electrical conductivity.

Although I decided to test the transformer and chokes before proceeding they seemed to be OK from a physical examination, so I took a chance and fitted the mains transformer to the chassis, fitting a tagstrip for the high voltage connections. This is essential in order to prevent accidental breakages of the ancient wiring and to anchor the leads prior to electrical testing. Fitting the transformer to the chassis made the job of testing it a lot safer.

The next step is to trace the transformer connections. From the original mains tapping panel it appears to have a 240 volt winding with various tappings.

The HT output looks like a single centre-tapped winding, and there appears to be a connection to an internal screen plus four low voltage windings.

Tag	A	B	C	D	E	F	G	H	I
A	x	9.2	3.4	9.7	2.4	x	x	x	x
B	9.2	x	11.7	1.3	10.8	x	x	x	x
C	3.3	11.5	x	12.2	1.9	x	x	x	x
D	9.6	1.2	12.1	x	11.1	x	x	x	x
E	2.3	10.7	1.8	11.1	x	x	x	x	x
F	x	x	x	x	x	x	x	x	x
G	x	x	x	x	x	x	x	463	223
H	x	x	x	x	x	x	463	x	238
I	x	x	x	x	x	x	223	238	x

I labelled the tagstrip connections A-I and from resistance measurements in ohms, I reproduced the transformer winding connections opposite. From the mains tapping panel B-D is 10V. A=240V, B =230V, C=220V. HT & LT will be measured once testing is underway. There was no measurable leak between the screen, F and the windings.

I painted the chokes and transformers in a matt black but decided to leave the extra choke (LFC3) as it's in passable condition.

Above, the main chassis with freshly painted parts temporarily in place. It will be easier to fit ancillary parts such as valveholders without these heavy components bolted down. The design, as modified by the previous owner used a pair of KT66 valves (one of which was intact) plus a pair of EF37A pentodes (both in place). Probably a rectifier such as a 5U4 or 5R4 was used? Also in place was an aluminium outrigger chassis, screwed to the rear of the main chassis, carrying LFC3 and a large condenser, C11. Because the amplifier relied totally on the receiver for its input level it seems sensible to add a master volume control to this chassis, together with a mains on/off switch (previously the one on the receiver chassis was responsible for turning on the amplifier/power supply).

I decided to put back the outrigger chassis to carry the extra parts of LFC3 and C11. No rust of course because it's made from aluminium.. and as you can see a piece of much used aluminium. It seemed a shame to use a fresh piece and lose a bit of the amplifier's history.

Not easy to see here but it's cleaned up and now painted white to match the chassis. The (old valveholder?) holes will be covered up by the choke and condenser

I also decided to repaint the casing of LFC3 which was a maroon colour but very rusty.

When these pieces are detached you need to be careful not to dislodge the core laminations. The end block of these can just fall off as often (and in this case) these are not interwoven with the remainder.

Below.. slowly getting reassembled. The KT66 IO sockets needed their fixing holes filing because the original B4 sockets had wider fixings. The last owner had fitted ceramic IO bases fitted with securing springs. The mains panel and fuses cleaned up OK.

Above; was this intended to be the front or rear? As the outrigger chassis was fitted to the other side, I shall refer to this as the front. The second and third holes in the front of the chassis accommodate the power and audio cables connecting to the receiver. The designers made a fundamental error here. The cables from the amplifiers carrying HT and LT have male plugs which risk a short-circuit if unplugged from the receiver. The first hole carries the mains lead and the fourth a three pin socket for the single loudspeaker (of course this amplifier is mono not stereo).

Good news... I've tested the mains transformer and it's fine. I also checked the three LFCs and again all tested OK so the amplifier rebuild should be able to proceed without risk. The mains selector panel was interesting. The two plug-in jumpers were open circuit at multimeter test currents (although possibly OK at real values?). The pair of pins is bridged by a metal plate whose tarnishing produced an open circuit. The solution was easy. I added a wire links across the pins of the two jumpers, soldering to open ends of the pins. This preserves component originality.

The mains transformer produced open circuit (no load) voltages of 6.5V, 5.5V, 4.5V, 4.5V and 470V-0-470V. The incoming mains measured 240 Volts. I was interested to see the HT voltage given a nominal 100mA or so load so I used two 1.8Kohm wirewound resistors in series as a dummy load and read a voltage of 432 volts RMS across half the HT winding. The current through the load measured 120mA. This represents around 52 watts or over 100 watts once full wave rectification is used for the full winding. I checked the chokes and found these had DC resistances of 500 ohms for two and 300 ohms for the largest. Total voltage loss at LFC1 for the KT66s running say 40 watts input will be about 30 volts, but add to this a receiver and phase-splitter drain of say 40mA, increases this to 40 volts making the KT66 anode supply voltage say 390 volts. Further down the HT chain the receiver HT will be circa 370 volts and finally the preamp HT after LFC3 will be perhaps 360 volts. These figures are rough and ready because the rectifier output voltage will depend not only on the transformer output but also on the reservoir condenser and anode resistance in the rectifier.

The three smoothed HT supplies will be used for the KT66 anode and screen supply, the receiver power supply and the amplifier EF37A phase splitter. When first purchased the amplifier probably used a pair of PX4 valves which each use a 4 Volt 1 Amp filament supply. This transformer winding is now redundant. The receiver has two heater supply feeds for some yet unaccounted for reason, but checking the low tension transformer windings appears to indicate there is only a single 6.5 volt winding so why should a pair of receiver valves use only a 4 to 5 volts heater supply? One of the low voltage windings is used for the HT rectifier, no doubt that closest to 5 volts for a 5U4 or a 5R4, either one capable of running over 200mA, although voltage-wise the 5R4 would appear to be better considering the very high off-load HT voltage which will probably be in excess of 500 volts. The rating of the old condensers looks OK as long as they have not degraded to the point where they draw excessive leakage current.

Above... the main components screwed down to the chassis excepting the outrigger chassis which will be fitted later because it makes handling the chassis for wiring more difficult. Safety-wise, this receiver-amplifier package, aimed at the home experimenter (in view of the sales of chassis without cabinets) would have been an extremely dangerous proposition, certainly not marketable in the present day.

I've yet to decide on how to run the two KT66 valves. The original amplifier most likely used a pair of PX4 triodes. The last owner swapped these for a pair of KT66s but not triode-connected. Considering that there are a couple of spare LT voltages available I might use these to establish a negative bias supply and run the KT66s at up to 50 watts output. Note that I've incorrectly referred to the pair of EF37A valves as a phase splitter but that is not the case. Phase splitting is carried out within the receiver where one valve (V10) provides an undoctored audio output whilst a set of valves (V11, V12 & V13) driven by V10 provide an audio output carrying treble and bass boost. V9 in the receiver is the audio amplifier fed by the receiver circuitry and the gram input. V9 happens to be fed from its own heater supply which seems to me, after a quick test to be circa 5 volts (might this have originally intended to have been rectified to feed V9 with 6 volts DC?).

All this means there's a problem in making this amplifier into a general purpose equipment because it needs a pair of anti-phase audio inputs. Would it be possible to arrange a switch enabling input from either a Morton Cheyney Receiver or a standard audio source? I'll need to sketch out the implications before I get too far in the rebuild programme. Already, I've decided to incorporate a volume control so this will be added into the design process...

Another factor to consider is swapping one of the EF37A valves (V2) to an ECC32 to replace V2 & V3 and using the other EF37A (V3) as a phase-splitter for a standard input.

The original circuit (or at least the circuit I traced on the chassis.. left) has the two EF37As wired as pentodes.

And then there's the option of changing the output circuit back to triodes by strapping the KT66 screens to their anodes.

I should really look at T2 to determine exactly what its characteristics are. Was it designed for the PX4s or is it a replacement designed for KT66s?

R1	100 x 5W	R7	330K x 0.5W	R13	10K x 1W	R19	100K x 1W
R2	100 x 5W	R8	330K x 0.5W	R14	10K x 1W	R20	240K x 1W
R3	10K x 1W	R9	500 x 10W	R15	33K x 5W	R21	120K x 0.5W
R4	10K x 1W	R10	51 x 1W	R16	47K x 5W	R22	100K x 5W
R5	240K x 0.5W	R11	51 x 0.5W	R17	250 x 1W	R23	2.2K x 1W
R6	240K x 0.5W	R12	500 x 10W	R18	120K x 0.5W	R24	6.2K x 1W

This is the audio section of the receiver

The next task is to look at the output transformer. Three things are necessary.. first, all the HT stranded wiring is insulated in badly perished rubber which needs replacing (the low impedance outputs are single wires insulated in a cloth based insulation which is OK). Secondly I need to check the HT wires to see what exactly they are (there are a total of seven wires) and thirdly I need to establish the various transformer ratios and impedances to see if the transformer is designed for KT66s or the original valves (whateverer these were... PX4/PX25 or DO41)

I removed the transformer shrouds and cut away the insulated sleeving which is used to group the various sets of wires together and pulled off the decayed rubber from the wires. I used heatshrink sleeving to re-insulate the wires then refitted the shrouds and having labelled them A to K, checked the resistance (in ohms) between the seven HT wires and other wires to determine the various connections, just as I'd done with the mains transformer. During the work above I found the only primary wires carrying solder were transformer connections A-C-B.

Wire

182.6

87.7

7.9

173.6

11.3

196.2

182.6

95.0

175.2

9.0

193.7

13.7

87.7

95.0

80.2

86.1

98.9

108.6

0.5

1.6

7.7

174.9

80.2

166.0

18.9

188.6

173.6

9.0

86.0

166.0

184.8

22.7

0.5

11.2

193.7

98.8

19.0

184.8

207.4

196.2

13.7

108.6

188.5

22.7

207.4

Left is the output transformer wiring diagram based on the resistance readings above. It seems to cater for three different types or modes of operation of output valves and three (or more) speaker impedances.

More testing is needed to identify the matching speaker impedances.

Next, I fed 4 volts at 1KHz into winding F-G (then D-E) and measured the outputs at each full primary winding in turn. Comparing F-G with D-E and J-K showed D-E and J-K were almost identical and were a factor of 2.2 different. Ratio 1 gives the step down ratio from primary to F-G and Ratio 2 primary to D-E.

F-G/D-E	H-I	F-G/D-E	A-B	F-G/D-E	L-M
In	Out	In	Out	In	Out
4V	100V	4V	115V	4V	129V
Ratio 1	25:1	Ratio 1	29:1	Ratio 1	32:1
Ratio 2	55:1	Ratio 2	64:1	Ratio 2	70:1

I then altered the test frequency and found much the same results from 200Hz up to around 10KHz. The true bandwidth may be better than this because my signal generator wasn't a good match into the low impedance transformer winding outside this range. Per valve, the ratios will be halved as I ignored connection C and measured across the full winding.

In summary the transformer appears to be very flexible because of its variety of connections. For a pair of triode connected KT66s with an HT of 400 volts its quoted anode load of 4Kohms can be matched into a 4 ohm loudspeaker connected across D-E by the calculation: Ratio squared times 4 ohms, which gives from the table above, (half of 64 because each valve sees half the primary winding) 32 squared = 1024 x 4 ohms roughly equals 4Kohm. So winding A-B (with the tell-tale solder on the wire ends) seems to be correct for a match into a 4 ohm speaker. For 8 ohms I would connect the two windings D-E and J-K in series because these windings measured as equal. Now that there's no risk of shorting the winding connections, I'll refit the transformer and try testing using KT66s. As an aside I've copied a circuit of the Williamson Amplifier below which dates from the same period as the Moreton Cheyney and I wouldn't be surprised if this wasn't spotted in Wireless World to form the basis of the current design. In fact, this is the design I shall be adopting in the rebuild as it's very close to the circuit I dismantled apart from the fact it uses triode connections for the KT66s and triode drivers. I quite like the latter as it avoids carrying grid connections through the chassis. I'm using a balancing pot (R12) between R11/R13 which was used in an earlier design.

Not yet functional, but a quick check to make sure everything actually fitted. Wiring isn't completed and final design of the input area not finalised.. ECC31, 32 or 33? The latter two have a higher maximum anode voltage. There's also 6SN7, 6SL7, ECC34 and ECC35 not to mention reverting to EF37A... Above, the L63 was used. This is essentially a 6J5 which is doubled up in the 6SN7. Below.. all the valves are new old stock.

I tested the block condensers before wiring into the circuit. All three plus the one for the outrigger chassis proved to be in near perfect condition when I applied 500 volts across their terminals. I also re-tested the mains transformer. It must be the first one I've tested that had no hum whatsoever, but under no load. Primary input measured 247 volts into the 240 volt tapping; HT read 484-0-484 volts and low voltage secondaries 6.75, 4.67, 4.67 and 5.70 volts.

I spent most of the day wiring the chassis in accordance with the Williamson amplifier circuit above. I decided to use modern resistors as these are more compact and obviously more reliable then old carbon types. I thought carefully about condensers. I have a bag of 0.22uF x 1000V about 40 years old so checked these for leakage before selecting a pair for C6 and C7. For C3 and C4 I used a pair of 68nF yellow plastic types rated at 250VAC of about the same vintage. Fitting the three potentiometers, R12/R17/R21 was slightly problematical because I don't want to drill the chassis or make mechanical design changes. Although two are specified as 100ohm I had a bag of small 200ohm 2 Watt pots that were easy to mount on tagstrips so I used these. I have a very large number of old PC power supplies which provide an excellent source of connecting wire in various colours and of adequate voltage rating.

Once I've completed the wiring of the amplifier I'll fit the triodes and carry out tests to see if the results stack up against the figures appended to the circuit diagram. One variable is the HT supply. Initially I'll use my variable HT bench supply to see if performance changes as the HT is higher or lower than shown in the circuit diagram. Temporarily I'm using a 500 ohm wirewound ballast resistor fed from LFC2/C4 (on my PSU circuit) and I'll be feeding the variable supply into LFC1/C2 before I wire up the rectifier, a new U52 whose box is marked U52/5U4.

First phase of testing went OK except I found the gain was miles higher than the Williamson figures. I converted their peak (shown in the circuit diagram) to RMS and the results are shown below. The "max" values are just before clipping was visible. My HT was about 417V rather than 450V but not much visible change occurred after around 300V. Current consumption is 24mA with the two 6SN7 valves in place. Initially I found unbalanced outputs of 40/50Volts at around 200mV input but, after puzzling and adding a parallel 56Kohm across R7, I realised I hadn't wired in C2 or C1 (C2 is necessary as it defines the audio voltage at C3 because without it the anode load for AC is R6+R7 not just R7 as the design dictates) I used two new 6SN7s for V1/V2 and V3/V4. C2 fixed the unbalance but I still need to fit C1 which will reduce the gain of V1 somewhat. I used a pair of 33Kohm resistors for R11 and R13 (instead of 39Kohm) and a 10Kohm pot for R12 (instead of 25Kohm) so my gain should have been lower than that of the original Williamson circuit by around 2.5dB. Up to now I haven't wired the feedback loop, hence the extra 26dB gain.

Test at 1KHz	Williamson	Gain	M/C nominal	Gain	M/C max	Gain
Input RMS V1g	1.34V		0.1V		0.2V
Output V3a	26.9V	22dB	26V	48dB	52V	48dB
Output V4a	26.9V	22dB	26V	48dB	52V	48dB

Before I add the two KT66s and check their output I need to finish some of the chores.. such as tidying up the wiring. I also need to make a couple of temporary panels secured to the ends of the chassis on which to balance the amplifier upside down (below). This is because the KT66s are taller than some of the surrounding parts and I need to add supports to protect them from damage. This done I added a mains lead connected to a switch on the rear of a volume control which I added in place of the fixed 1Mohm input resistor (a 2Mohm Radiospares volume control with a log characteristic). The M.C. designers fitted the loudspeaker output socket next to the first audio amplifier but hopefully this won't result in feedback problems. Another rather odd design feature was the positioning of the output transformer between the amplifier and output valves. This means that the leads connecting the coupling condensers are rather long and I must be prepared to make wiring modifications if I'm troubled with hum from earth loops. I fitted a pair of 1uF 600V decoupling condensers for C2 and C5 and a 0.22uF 1000V condenser at C1.

Further testing will proceed with an external variable HT supply as before, although I plan to use a pair of HT supplies to provide more flexible testing and sufficient current.

With the KT66s plugged in I connected an HT supply to the driver stages and measured about 24V RMS at the input ends of C6 and C7. I then connected a dummy load of 8 ohms across the first low impedance winding of the output transformer, applied HT to the KT66s and cranked it up. At 400V they were drawing 100mA and the output voltage measured 11.3V. Trying the other two windings gave me about 5.4V. I reckon this works out at a little under 16Watts output with an input of 40Watts. The feedback loop (R25 to R4) isn't connected yet. Below some pictures taken during testing.

1.Above left: Input 292mV showing the input to the amplifier at C6.

2 Above: Matching anti-phase signals at KT66 inputs C6/C7.

3. Left: C6 voltage with reduced input 119mV

4. Bottom left, KT66 output into 8 ohm load with C9 at 24.1V (8.7W).

5. Below: KT66 output into 8 ohm load with C9 at 30.8V (15.7W)

Notes:-

Feedback loop not yet connected.

C10/R26 not fitted.

R21 and R17 not yet adjusted for KT66 balance.

You can see the start of flat topping in the 16 Watt output (last) picture. This is probably connected with the voltage (re. 90V Peak) at the anode or cathode of V2 or the anode of V3 or V4. I'll work out which and see what can be done. The output sinewave looks tilted to the left. This might be because R21 or R17 isn't yet set up correctly.

On the right a picture which explains a few points. Having first adjusted the dummy load to 4ohms I noted the output voltage dropped as expected. The shape of the waveform was reasonable in shape and I decided to fit the feedback resistor. This is given in the Williamson circuit as 1,200 x the square root of the speaker impedance which works out at 2.4Kohm. When this resistor was fitted I found the waveform improved to the extent that, visually at least, it looks distortion free. The output here is (5.1 x 5.1)/4 or 6.5Watts. The main point is that, with the feedback resistor in place the input necessary for a given output has increased from around 100mV to close to a volt which more closely matches the figures in the Williamson circuit diagram. Those figures give the input as 1340mV for an output of 15Watts which suggests my choice of 4ohm output winding is wrong.

The KT66 anode is sitting at 325 volts with its cathode at 30 volts. The current is shown on the HV PSU milliameter as 130mA making the input around (325-30) x 0.13= 38Watts so efficiency isn't too good.

If I increase the input voltage the output begins to distort.

There are two other output transformer speaker windings (both having the same DC resistance of 1.3ohms) so the next step is to test one of those. My guess is that these may be 4 + 4 ohms? I'm using winding J-K which according to the tests should be a higher impedance than D-E. Because D-E is the same as F-G logic suggests these are both 4 ohms, making 8 ohms when connected in series with J-K perhaps 16 ohms? There are also two other primary windings to consider, Currently I'm using the middle one, A-B. The choice really depends on the final KT66 anode voltage/current draw and that essentially is determined by the mains transformer and as I'm using an external HT PSU which has a max output of 400V x 100mA the anode load is as yet not to spec.

I spent most of the day tidying the workshop then decided to carry out a spot more testing. I used the winding (1.3ohms) adjacent to the higher impedance one (2.7ohms) and sure enough the output was somewhat higher as expected. Initially the amplifier went beserk but then I realised the wires were soldered the opposite way around and instead of negative feedback I'd applied positive feedback. I switched the connections and the amplifier became stable again. The maximum undistorted power into the dummy load at 1KHz was now (6.48V x 6.48V)/4ohms=10.49 Watts. I then altered the frequency of the drive as shown in the table below, adjusting the gain control to achieve an undistorted waveform at each frequency change. Low frequencies were prone to distortion beyond 850mV input and at the higher frequencies there was tiny bit of ringing on the waveform indicating maybe some minor changes are needed (for example increasing the capacity of the coupling condensers) although none of the test leads was screened which might account for this. The dummy load was a large wirewound rheostat which will have some inherent frequency response and I was using an 0.05uF input coupling condenser between the audio signal generator and V1 to reduce damping (160Kohm @20Hz and 160ohms @ 20KHz) hence the need to keep adjusting the input pot etc. A fixed resistor would be better.

Frequency	20Hz	50Hz	200Hz	500Hz	1KHz	2KHz	5KHz	10KHz	20KHz	25KHz
Input	832mV	1.26V	1.26V	1.33V	1.22V	1.62V	1.8V	1.84V	1.71V	1.89V
Output	4.44V	6.29V	6.47V	6.9V	6.48V	8.5V	9.24V	9.4V	8.7V	9.55V
Power in 4ohms	4.9W	9.9W	10.5W	11.9W	10.5W	18.1W	21.3W	22.1W	18.9W	22.8W

Ideally I'd like to use a spectrum analyser with a tracking generator to check the response, before I did this I decided to use the internal power supply rather than external HT supplies. I used a 5U4 valve as a full wave rectifier and fitted a 100 ohm resistor in the grounding wire from the HT centre-tap of the transformer plus another in the HT feed to the preamp. The various readings at max undistorted output into 4 ohms measured as follows...KT66 anode voltage 425 volts with 10 volts across the HT current monitor, making 100mA. Each KT66 cathode has about 300 ohms to ground via their self-bias resistors so these produce a bias voltage of around 12 volts. Preamp current measured 2 volts at 100 ohms = 20mA with the HT at 412 volts. Output voltage measured 8.6 volts RMS across 4 ohms = 18.5 watts RMS output . Input power measured 425V x (100-20)mA minus the effect of the grid bias network (12 volts) = 33 watts. I found I could increase the drive and the output rose but began to get distorted. The efficiency of the KT66s works out at about 56%. To improve this I guess I could connect the KT66 grid leaks to the negative voltage at the transformer HT centre-tap making a total of minus 24 volts (if the output parameters remained the same)?

Looking at the results from previous tests, the enhanced HT voltage has resulted in a power increase from 10.5 watts to 18.5 watts at the test frequency of 1KHz, but of course I'm unsure about the amount of distortion I'm getting until I do further tests.

pending

Rebuilding the Moreton Cheyney Amplifier

Return to Reception