Home Tubes Asylum

Questions about tubes and gear that glows. FAQ

Complete Retube of an Audio Research VT100

I turned on my VT100 for some normal listening and noticed that the center image had shifted to the left about 3/4 of the way. I swapped speaker cables, and the image moved to the other side. I swapped interconnects to the amp and nothing changed. I then knew that something was probably wrong with the amp.

I swapped the left and right quads of output tubes that I had recently replaced. The sound did not change. I then, with great hesitation, swapped the small driver tubes. The image shifted to the right about 3/4 of the way. I swapped them back, the image shifted back to the original position (Skewed to the left). Now in addition there was a microphonic squeal in the right channel (the weak channel) that sounded like a guitar amp with a bad tube or a Conrad Johnson PV series preamp with a microphonic tube. I immediately shut the amp down.

I had been lucky enough to stumble across Abe Collin's site on VT100 retubing so I felt like I could handle the retube myself. Before reading ahead, you make want to check out his page or a lot of this won't make sense. Here is the link:

I called Upscale Audio and got 12 tubes sent to me (8 for the amp, 4 for the preamp). I called Audio Research and they faxed me instructions and a schematic. I then aquired two Fluke multimeters and a set of clip adaptors from Radio Shack to latch onto the small parts while setting the voltages.

I got the tubes and tried my best to group them closely into two sets of quads. According to the instructions, you should remove the output tubes when adjusting the amp, but you should never turn on the amp without the preamp tubes. Also, you can run the VT100 without a set of speakers attached.

I installed the quads and turned on the amp. I attached leads as shown in the photos on the Collins site. I let the amp warm up for about 20 minutes. I adjusted the driver tubes until they tracked within .001 volt of each other. The is not as easy as sounds. Each tube reacts to the other and getting them together takes a little trial and error, but eventually you will learn to bring one up too high and then use the other to bring it down and the two tubes will get closer to unity. Even worse, the tubes drift and are sometimes slow to respond to small adjustments. This won't make sense until you see it done. I set them each to around .100 volts and then checked the +60V on the board. I then adjusted each driver tube half up or down until the +60V reading had gone half the distance to the 60V spec and then rechecked their balance with each other. Once I had the two driver tubes withing .001 volts of each other and the +60V within .01, I checked the balance between the two Infinicaps.

The spec on these has to be within 20V of each other and around 160V. I ended up with 22V apart on the left and 30V apart on the right. The only thing you can do is power the amp down, swap some tubes, and start the process over. Any changes will affect the balance between the driver tubes, the +60V spec, and the balance between the infinicaps. After many hours and many changes I had the left channel within .03 volts of each other and the right channel within 2 volts of each other. I put all of the covers back in place, let the amp run for another hour, and then checked everything again. I made sure that the drivers were at unity, I made sure the +60V on the left and right channel matched, and I made sure that the voltage on the two infinicaps was still close to the same. Everything stayed in spec.

Unfortunately, getting the tubes in the right places to make those two infinicaps match up was pure luck. There may be a method to the driver tube placement, but I unfortuantely did not keep up with which tubes came from which boxes that they were shipped in. The boxes had the triode measurements that would have told me a lot about the final order. Maybe I will keep up with them in 5-10 years when its time to do this again and develop a method to save a lot of time.

Anyhow, I replaced the output tubes, warmed up the amp, and biased the output to 130 on each side.

The sound was wonderful.

I learned a few things about retubing this amp. First off, if you are not sure of your skills with a multimeter, being able to identify components, being able to read a circuit diagram, and patience then don't try this job. You will need AT LEAST TWO digital multimeters, insulated clip adaptors, steady hands, and a LOT of patience.

This was not an easy job and half-assing it could leave you with a broken amp or worse.

In my case the time invested was well worth it. The NOS Sovtek 6922 type 2's did a wonderful job. Vocals are now much more detailed. The soundstage is as good or better than it was before. The highs are better. The bass still kicks ass, maybe better than before. Best of all, I was able to do this myself with a little help from Abe Collins' website, Leonard at Audio Research, and Kevin at Upscale Audio. I want to thank all of them.


This post is made possible by the generous support of people like you and our sponsors:
  Michael Percy Audio  

Topic - Complete Retube of an Audio Research VT100 - SAT 15:39:02 08/12/02 (5)

You can not post to an archived thread.