Rebuild of Dynaco fm-1

See last page for enhancements and "bugfixes"

It started with a need for buffering my PAS-3 output. As there is not enough power available to add a cathode follower,an external device of some kind was needed. In addition to a power supply a suitable box was needed, this started a line of investigation. Another thing i'm missing is a way to reduce / increase volume while sitting down.. This is nothing new for many people but i am old and my PAS-3 has no remote:-(

Initially i was thinking of a potentiometer rotated by a servo of some kind, an arduino microcontroller was an obvious device to control this.

Thinking about casing i came to think in another thread, the fm-radio as we know is soon killed by clueless administrators and replaced by DAB. Thus any fm-tuner will soon be a historical piece of junk. Well, why not use a Dynaco fm-1 as donor for my case and power supply. Some metal-work and an IEC well is mounted in the back. ( fm-1 was really never a good tuner anyway )

While testing various potentiometers for adapting a servo i stumbled across this ,clearly interesting and cheap too. A quick discussion with ebay and the device was "in the air".

When the pot arrived i had to ask for additional documentation ( there was none) which i received in decent shape. Connecting to a DC source and installing a battery in the remote and i was convinced. This is what i need.

The original arduino still has it place; i will use it for power sequencing and the possibility to power on/off by the remote. I already have an SSR in my VTA-70, i will install one in the fm-1 and one in my PAS-3. Those will be controlled from the arduino, but with a twist:

The VTA-70 has a rectifier tube ( still). It does not like turning on again shortly after it has been shut off, this might cause power surge that causes flashing of rectifier and / or power tubes. Thus the arduio will have a software delay to cope with this.

First pictures taken during initial testing

Front panel view, the remote-controlled remote is mounted at the tuning control's place. The indicator is replaced by a display and the remote IR-sensor. The "volume" pot is replaced by a stereo pot and will be used for adjusting the "tape-out" level ( there is 2 catode-followers inside, one for the poweramp and one for the tape-out). The tape-out is connected to a mac-mini and used for ripping vinyls.


Front view , well it's work in progress

A first glimps of the innards, it's work in progress, arduino and SSR's not installed yet.
The digital side is powered by it's own transformer, this is intended to remain powered on ( controlled by the front-panel power switch ). A small regulator ( from china/ebay) smoothens the 7V DC used for arduino and remote.

To the left some caps may bee seen, i had to mount them on top of the chassie as the bottom side is to narrow. But all resistors remain "below deck" as usual.

In the back there was space enough to enlargen the power inlet in oder to install an IEC power well. This ought to be "standard upgrade item", i cannot feel confident about the american habit of "a knot on the powercord".

A "birds-eye" view. The empty tube is the second cathode follower for the tape-out. Ignore the loose cables ..

Here is the arduino and relay module ready for mounting. Note the "shield" mounted on top of the arduino, its where signal conditioning is done. Shield is the term used in the arduino community, other people would call it "interface board". Here a prototype shield is used to mount everything needed to interface outside signal, it's both an issue of signal integrity and protection of the arduinos ports. One signal is 24V DC ( power on from PAS-3) , the other signal is a open collector from the remote controller. In addition resistors in series with outputs is needed to reduce current to safe levels. By mounting all this on a "shield" the arduino is unmodified and easily replaced if needed.

Now playing BACH

All electrical parts mounted and working. Backside of cover remains, so does some panel fastenings.
Leftmost is two JJ ECC82 ( or 12au7 for the overseas ). Then comes input and output caps mounted above deck as the underside is limited in space. Then comes the original power supply, the only difference is that the filament is elevated to 1/3 of B+.

Top right is the arduino controller, bottom right the DC supply and regulator for +7V and, behind the knob, the chinese remote volume control. There is no button on the tape-output pot.
Bottom view To the left is the underside of the arduino controller, dual relay board and a regulator for +5V is seen. At top two rca-connectors are mounted, the leftmost is the delayed output that controls the poweramp, the right one is intened for the preamp.

The black square tilted 45 degrees is a sharp S202S12 SSR. It needs very little cooling as the power consumed is a few watts. It's not as convenient as the original IR SSR, as it need an external resistor to limit current, this resistor ( 470 ohm) is mounted right above.

In the middle is the remnants of the fm-1 power supply, quite standard except for the IEC input.

to the right is the two ECC82 mounted, so is all resistors used, all carbon comp (!)

Remaining work, or problems if you like

I'm haunted by hum ! At least i have 50hz hum at -66dB, 100hz at -70dB. As i have not mounted the covers and all in/out cables lies in a bug pile behind i will postpone this. The only thing i will do is measure the B+ ripple, and if excessive find a replacement for the B+ cap(s). This is unlikly as the 50hz is significantly greater then 100hz, this tells me that i have AC injection hum.

The idea of 2 ways of turning on/off ( either via DC signal from preamp when that has been switched on, or by detecting a signal from the remote ) will be scrapped. I'll settle with the preamp DC only. The previous solution was complicated and does not reflect real usage. I'll go for a better solution. I will do this by simply lifting the cable from the remote in case i want it later. I will however create another software feature : if the arduino comes to life AND the pre signals "ON", then i will not proceed starting the amp, instead i will signal this as an "power glitch". To reset this, simply turn off the pre, and then on again. This will protect the poweramp from restart in the event of power fail.

Hum issue resolved

A missed ground connection on a cable to the tape-out buffer found. Now 50hz hum is at -78dB on the tape-out amp.


Now playing..

Mounting covers to see if everything fits.


I also digged up and mounted 4 rubber feet's, chopped off the tape-out potentiometer and mounted the knob.


Now my PAS-3X seems dirty and dull. Well, i don't care for the moment. Now playing "tottas bluesband" that is more importent then some dusty front panel.

Software update

Re-reading dave gillespies article about "power tube flashing and the reasons" got me adjusting the arduino software.
Function is as :
- at startsignal(1) it will delay 3s before powering up preamp. This prevents preamp powersurge if power fluctuates.
- after startsignal it will power on poweramp after 25s.
The above delays are unconditional, under no conditions will amps be "power cycled " faster then this. If power is dropped the startcycle will be repeated from beginning. When startsignal is removed the poweramp is shut down. After 3s will the preamp be turned off. This will give poweramp time to unload it's caps and thus be safe even if the preamp will surge it's output at power off. If powerdown cycle is aborted and a new startsignal is detected, a normal startcycle is done.

(1) startsignal is the signal from preamp or remote to start poweron cycle.
Due to request, the arduino code FM1.ino

metrics

Power consumption at 220V : 2w at standby, 26w active.

Failure!

Strange sounds, like fried bacon comes from my speakers.

Problem was the 18k cathode resistors, i had not done my homework but installed 1/2w . As used they are burning at 1.1w!
Install new ones, and mount them at the edges to give the new ones better cooling.

Everything else looks good, the JJ-ECC82 ( delivered by tubular-well.se , my own tubedistributor) measures as new, ( they should, they are only used 200h).

I have received question about the cathode follower, local copy : schematic,
it's a direct copy from the excellent : valvewizard.co.uk , bottom of page.

Due to request, the arduino code FM1.ino

Thanks for watching this, your comments are much valuable! Edited 2015-03-26