Guitar amp conversion
1947 Philips 2864
The "ancestor" of the Tube WorkShop amps. When it got to us it was actually just a heap of junk!
In the end, this became the "blueprint" for two more TWS amps...
This amp probably saw the light of day around 1947.
A then highly modern music amplifier that could switch back and forth between 3 different input sources: "Microfoon", "Gramofoon"; and "Radio". This is what the Dutch documentation for the amp reveals, which can be found at www.radiomuseum.org. By the way, you can also find a circuit diagram here, which unfortunately contained an error. I corrected it and uploaded the corrected version there..
In the course of time, several electronic artists have probably tried to make a statement by doing modifications to this amp and I was expecting a non-functional "wire-hog" that no longer had too much to do with the original circuit.
In terms of looks, the Philips 2864 is more... well, let's say "rustic" and "simple". But ok, it was 1947... just after WWII. You didn't have anything... But wait, you had one thing back then: Tubes with secondary emission!
And that's why we don't start with the housing, power supply or power amplifier within this amp, but we directly go for the exciting preamp.
The Preamp
with the mysterious EEP-1 tube
"Secondary emission"... a term that I had come across as an undesirable side effect in tubes.
"A secondary emission can sometimes occur because the anode is bombarded with too many electrons, causing it to emit electrons into the tube, which are then picked up by the screen grid."
According to the literature...
Not everyone has to understand that... Fact is: We don't actually want that, because it normally degrades the performance of the tube. Therefore, over time, all kinds of suppression grids were introduced into tubes to minimise this effect.
But there are a handful of old tubes that use exactly this effect, provoke and amplify it in order to generate a second, phase-inverted output. In other words: An attempt was made to integrate a phase inverter directly into the preamp tube via secondary emission. One of them is the EE-1 or EEP-1 tube that is used in the Philips 2864.
Let's say that should be used in the 2864...
In our amp, a very motivated hobbyist had replaced the EEP-1 based preamp in the past with a combination of an EF40 pentode and an ECC40 double triode. Accordingly, the housing opening for the P8A tube socket of the EEP-1 was roughly widened with a file and an 8-pin or Noval socket was installed. So we got 2 great old ones, Philips EF40 and ECC40... but no EEP-1. Damn it!
I briefly thought about analyzing and understanding the built-in circuit, but discarded this after about 10s. My goal was to use the original circuit of the amp and adapt it to guitar. So that meant removing the mutant preamp completely and then rebuilding it.
Conclusion: The EEP-1 has to get back into the preamp.
This tube has an insanely high gain of >100 (if you really like to push it) with a simultaneously integrated phase inverter through the secondary emission. A true "Tube from Hell". Everything was packed in there that possibly could have been packed in at the time.
What is this telling to us? This tube is a sensitive one and will be very susceptible to microphonics and mechanical shock (which it actually is).
In the pictures above you can see that the wiring of an EEP-1 involves significantly more effort than (for example) an ECC83, which only requires a handful of components. And by the way, it wasn't that easy to find EEP-1 tubes in good condition. In the meantime we have probably bought up everything that was still available in Europe. That was exactly 6 EE-1 or EEP-1 tubes. 3 of them still in NOS condition. Let's hope more are found in some dark basements and made available for purchase in the future.
But back to the 2864 preamp: We don't need 3 inputs for "Microfoon", "Gramofoon" and "Radio", only one for guitar with high impedance. For this purpose, the microphone input (that no longer existed) was converted to a 6,3mm jack socket and the other unnecessary connection terminals were shut down accordingly.
When dismantling the amp, it turned out that almost all original ERO capacitors were ready for the "Box of Shame" because they were already dissolving into their components. The original red "Dogbone" resistors, on the other hand, were still in top condition even after more than 70 years and thus went into the "Box of Fame" getting reused.
It was taken into account right from the beginning that in a guitar amplifier application there should be another amplifier stage before the EEP-1, which buffers the guitar signal and the volume potentiometer. Since we already got an EF40 with the amp (very similar to the EF86) with a corresponding socket, this was used in the further circuit design.
Although this tube (just like an EF86) has a very high gain factor in the pentode circuit, it was used very tamely in the 2864, using triode wiring and without cathode capacitor. It was not designed to deliver much "Gain", but stable impedance ratios for the EEP-1.
Regarding the optics, a P8A socket for the EEP-1 fitted well into the "carved-out" chassis opening (it was originally intended for this). The rough extension for the additional EF40 was covered with a half rubber grommet and serves as an opening for the grid connection of the EEP-1, which is done with a shielded cable and an attached cap. For the base of the EF40, a housing bore was used that was originally intended for a standing filter capacitor of the power supply. Since we will be replacing all filter capacitors anyway, we can accommodate them elsewhere.
So far so good. The combination of the EEP-1 with the EF-40 later turned out to be an excellent combination. I could now claim that this was of course absolutely foreseeable and meticulously planned by me.
Bullshit, sometimes you just have to be lucky in life... and things you try turn out to be a "perfect match"... ;-)
Power Supply
No compromise.
The mains transformer looked a bit battered, but still delivered stable and good values in all respects. Especially the 4V heating voltage for the 1805 rectifier tube. A value that is no longer found today.
So, thank God, it could stay in the amp (along with the 1805 rectifier tube), as did the original choke inside the chassis.
The power supply electrolytic capacitors were mainly designed as standing types in the housing + a type attached internally by a clamp. It is clear that these electrolytic capacitors have to be replaced after more than 70 years.
There is always a question of conscience whether to leave the original electrolytic capacitors in the housing for optical reasons or remove them and replace them with modern types with a similar form factor.
In a completely untinkered and original amp I tend to the former, as long as there are no mechanical 1:1 replicas. I find it very difficult to mechanically change an amp in its absolutely original condition and to mill openings or drill holes for clamps for modern capacitors.
With our present 2864, however, I decided to remove the original electrolytic capacitors and replace them with modern types with a clamp. A few smaller holes for the clamps had to be drilled for this, but due to the modifications in the past, the amp was far from being "original" anyways. So the focus was on the best possible function with current components without slavishly clinging to the "original look". Pure "Originality" was already lost within this amp anyways...
The filtering has been increased from the original 16uF to 32uF in most places, which leads to more stability and is still easily coped with by the 1805 rectifier tube. Special filter capacitors from TAD were used with a diameter of only 25mm that fit into the existing housing openings.
Power Amplifier
... needs work!
Looks pretty straight forward at first glance. 2x 4694 in push-pull and with cathode bias... output transformer... and "off for it".
The following problems:
1. The 4694 is practically no longer available, and the EL-3, which is often recommended as a 1:1 replacement on the internet, actually works with a significantly lower anode voltage than the approx. 380V in the 2864. Of course, our amp had EL-3s and none 4694s. It was kinda clear...
I managed to get a pair of 4694s that were in good condition and it turned out that they perform much better in the 2864 than the EL-3s. These always seem to struggle with the high anode voltage.
2. The original output transformer seemed to work at first. However, when he really had to transform power again, it became clear that the primary windings had partially lost their insulation and the output transformer's values therefore varied greatly depending on the voltage and temperature. Unfortunately, this could not be saved and ended up in the "Box of Shame".
An old output transformer from a 2xEL84 amp (same primary impedance) served as a replacement, which looked "historically correct", has a good number of laminations and allowed optimal impedance matching to 4 and 16 ohm speakers.
Tone Control
this can be done better.
In the original, the 2864 only has a "treble cut"; similar to a VOX AC30.
In general, I didn't want to implement an elaborate tone control that strongly influences the guitar signal, but rather work with the conditions of the 2864 circuit.
The function of the treble control could be adjusted much better using other component values, and the original selector switch for the input source proved to be the perfect platform for making the amplification in the bass range adjustable in the circuit around the EEP-1.
The combination of both controls does not allow a setting between "Fender Clean" and "Marshall Distortion", but it is an excellent tool for adapting the 2864 to the guitar and, above all, the type of box used.
Conclusion
a great accident.
We found an old Philips 2864 in pitiful condition.
The circuit had been modified several times and was not functional.
This little amp practically "begged" us to take care of it.
The result was excellent and unpredictable in this form.
After converting it into a guitar amp based on the original circuit, it quickly became clear that we had created an amp that sounds outstanding and far surpasses similar amps in terms of fine detail and 3-dimensionality. At least that's the opinion of people who have tested this amp.
Let's call it an"intriguing fluke" But hey, nothing ventured, nothing gained.
In any case, the revised Philips 2864 is the starting point for further Tube Workshop amps, where we concentrated on capturing the tonal properties of the converted Philips 2864 as authentically as possible...