Eico HF-87 Update

The old beast gets another life...


The oldest piece of audio gear I've got (in fact it's been around longer that almost everything else in my life, including my wife) is an old Eico HF-87.  It's been used and abused for decades.  Some ancient history on the amp is here.  It's been repaired and updated multiple times without any regard to collectibility.  In the 70's it was just another piece of old tube audio junk.  The time has come for a serious refurb.  HF-87's have got really good transformers and I've always liked the sound of EL-34s.  I'll update this page as the project moves forward which will depend on how much time I can find to work on it.

Here's what we're starting with:

Front of Amp Chassis
It's a bit of a mess and needs some serious cleanup.  The circuit board on top of the chassis is a FET driver board that we tacked on to test.  It will stay for the next iteration of this amp.  The current plan is to repackage the amp into something that be aesthetically compatible with other gear in a living room, refresh the fifty year old electrolytics in the power supply, integrate the FET driver board, and do it in a way that will allow the amp to be restored to the original HF-87 circuitry at some point in the future.  The amp will never have serious collectible value due to the mechanical hacks that have been done over time.

Project Goals
Make it reliable
Make it quiet
Improve the Wife Acceptance Factor (it's too ugly for the living room right now)
MAKE IT SOUND GREAT

Power Supplies
The B+ supply will remain as-is, with the old caps getting replaced.  The first stage caps, updated in 1978, are still working pretty well, and the P-P ripple is under three volts.  There's a 1H choke hiding behind the driver board, and that takes the ripple down to millivolts.  The bias supply caps will get replaced whether they need it or not.  (The power supply schematics are on the history page.)  And finally, I've been contemplating changing the filaments to DC to see if that helps quiet things down a bit.

FET Driver Board
This is an experiment.  A friend of mine builds amps using these FET input/driver boards and he swears by them.  We decided to drop one in the Eico to see how it would work, and so far I'm pretty impressed.  We didn't spend much time on the install though, and I think the "distributed grounding" that has occurred over time has created problems with hum and noise.  This will be cleaned up.

The Output Stage
Pretty much stays the same.

So as of 1/1/2010 that's the state of the project.  More updates after I get some parts in and get going.
Jim

Feb 2010 Update

The amps are as far as they're going to go for a while since I've been distracted by a set of Quicksilver monoblocks:

 Quicksilvers
Damned nice sounding amps, and they've got enough power to drive the old KEF's to reasonable volumes.

So the HF-87...

First, the power supply and grounds got cleaned up.  560 uF caps for the voltage doubler, and a 2 x 200 uF filter after the 1H choke.  Ripple is nearly unmeasureable with my scope - under a millivolt P-P and masked by noise.  Previously, the cap after the choke was an original electrolytic, and had lost of bit of it's capacity over fifty years.  (Who woulda figured?)  I was poking around with my scope while playing a CD on the bench and noticed a lot of audio - several volts worth - on this cap.  The choke passed the DC, blocked the AC, and with the lousy cap the supply was <really> soft.  Might account for the mushy bottom end...

I also replaced the "distributed ground scheme" with a 12 ga copper bus wire originating at the power suppy and setup a central star ground.  Turns out that the power supply ground on the lug of the transformer was actually loose.  Now everything that is supposed to be grounded is really grounded.  And the hum is nearly gone.

Chassis Grounding

So here's the end result:

Eico Amp Front

Which isn't nearly as pretty as I planned, but that's the job of the Quicksilvers now.  The important stuff:  How does it sound?  Really pretty good.  The hum is down to "normal" tube levels, which is to say it's not noticable once you back off the speakers and sit down to listen.  It's under the level of the turntable at listening levels, so I decided not to chase the DC filament idea, The sound changed significantly.  It's harder, faster, more analytical, and more accurate.  The bottom end is solid and tight.  Good detail.  Solid imaging.  On the downside, it's revealing and accurate enough to make some source material sound bad.  That tube softness is pretty well gone, and it's closer to a good solid state amp than I might like.

So now it's sitting down the basement parked on a Tabor speaker, looking lonely.  When I need the amp again, I just might put it back to the original topology with the 6SN7 phase splitters and premium caps and resistors.  And fix the Wife Acceptance Factor.  Now it's on to the ADC B100 preamps.